anil.recoil.org
+1
-3
aggregate_feeds.py
+1
-3
aggregate_feeds.py
······
+98
-18
eeg.xml
+98
-18
eeg.xml
···-<feed xml:lang="en" xmlns="http://www.w3.org/2005/Atom"><title>Atomic EEG</title><link href="https://example.com/" rel="alternate"></link><link href="https://example.com/eeg.xml" rel="self"></link><id>https://example.com/</id><updated>2025-04-10T01:00:00Z</updated><author><name>Feed Aggregator</name></author><subtitle>Aggregated Atom feeds</subtitle><entry><title>Dell R640 installation</title><link href="https://www.tunbury.org/dell-r640-installation/" rel="alternate"></link><published>2025-04-10T01:00:00Z</published><updated>2025-04-10T01:00:00Z</updated><author><name>MarkE</name></author><id>https://www.tunbury.org/dell-r640-installation</id><summary type="html"><p>Today we have racked the five 14th generation Dell R640 servers and a Dell N4032 switch.</p>+<feed xml:lang="en" xmlns="http://www.w3.org/2005/Atom"><title>Atomic EEG</title><link href="https://example.com/" rel="alternate"></link><link href="https://example.com/eeg.xml" rel="self"></link><id>https://example.com/</id><updated>2025-04-11T18:19:28Z</updated><author><name>Feed Aggregator</name></author><subtitle>Aggregated Atom feeds</subtitle><entry><title>Towards Global-scale Species Distribution Modelling</title><link href="https://watch.eeg.cl.cam.ac.uk/w/7aqBd2Nn9E6QpMvnoBPxuQ" rel="alternate"></link><published>2025-04-11T18:19:28Z</published><updated>2025-04-11T18:19:28Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/7aqBd2Nn9E6QpMvnoBPxuQ</id><summary type="html"><p>Abstract:<br />+Estimating the geographical range of a species from sparse observations is a challenging and important geospatial prediction problem. Given a set of locations where a species has been observed, the goal is to build a model to predict whether the species is present or absent at any location. This problem has a long history in ecology, but traditional methods struggle to take advantage of emerging large-scale crowdsourced datasets which can include tens of millions of observations of hundreds of thousands of species in addition to the availability of multi-modal data sources such as paired images and natural language descriptions. In this talk, I will present recent work from my group where we have developed deep learning-based solutions for estimating species' ranges from sparse presence-only data. I will also discuss some of the open challenges that exist in this space.</p>+Oisin Mac Aodha is a Reader in Machine Learning in the School of Informatics at the University of Edinburgh. He is also an ELLIS Scholar and former Turing Fellow. He obtained his PhD from University College London and was a postdoc at Caltech prior to his current role. His current research interests are in the areas of self-supervised learning, 3D vision, fine-grained learning, and human-in-the-loop learning. In addition, he works on questions related to AI for conservation and biodiversity monitoring. More information can be found on his website: <a href="https://homepages.inf.ed.ac.uk/omacaod" rel="noopener noreferrer" target="_blank">https://homepages.inf.ed.ac.uk/omacaod</a></p></summary><category term="EEG Seminars"></category></entry><entry><title>Dell R640 installation</title><link href="https://www.tunbury.org/dell-r640-installation/" rel="alternate"></link><published>2025-04-10T01:00:00Z</published><updated>2025-04-10T01:00:00Z</updated><author><name>MarkE</name></author><id>https://www.tunbury.org/dell-r640-installation</id><summary type="html"><p>Today we have racked the five 14th generation Dell R640 servers and a Dell N4032 switch.</p><p>When inspecting the rack rails, I noticed that some of the left-hand rails had an extra tab on them while the others did not. For the first server, I used a rail with a tab only to discover that the tab prohibited the server from being pushed in all the way. The tabs were easily removed but the server needed to be removed from the rack first.</p>···<p>To deal with folders of any size, we should use the marker system. For this, we pass <code class="language-plaintext highlighter-rouge">?usemarker=true</code> to the first GET request, which causes the API to return <code class="language-plaintext highlighter-rouge">next_marker</code> and <code class="language-plaintext highlighter-rouge">prev_marker</code> as required as additional JSON properties. Subsequent calls would use <code class="language-plaintext highlighter-rouge">?usemarker=true&amp;marker=XXX</code>. The end is detected by the absence of the <code class="language-plaintext highlighter-rouge">next_marker</code> when no more entries are available.</p>-<p>The project can be found on GitHub in <a href="https://github.com/mtelvers/ocaml-box-diff">mtelvers/ocaml-box-diff</a>.</p></summary><category term="Tunbury.ORG"></category></entry><entry><title>New preprint on benchmarking ultra-low power neural accelerators</title><link href="https://anil.recoil.org/news/2025-npu-bench-1" rel="alternate"></link><published>2025-03-28T00:00:00Z</published><updated>2025-03-28T00:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/news/2025-npu-bench-1</id><summary type="html"><p><a class="contact" href="https://profiles.imperial.ac.uk/joshua.millar22">Josh Millar</a> just released our latest preprint on how to make sense of the growing number of dedicated, ultra-low-power 'neural network accelerators' that are found in many modern embedded chipsets. My interest in this derives from wanting to decouple from the cloud when it comes to <a href="https://anil.recoil.org/projects/osmose">low-latency local environments</a>, and this needs fast tensor operations in hardware. Josh found a huge number of interesting NPUs in modern low-cost chips, ranging from <a href="https://www.espressif.com/en/products/socs/esp32">ESP32</a>-based boards over to <a href="https://arm.com">ARM</a> ones. All of these have quite a variety of tradeoffs, from the operations supported (which affects which models can be run on them) to the amount of memory and CPU power.</p>+<p>The project can be found on GitHub in <a href="https://github.com/mtelvers/ocaml-box-diff">mtelvers/ocaml-box-diff</a>.</p></summary><category term="Tunbury.ORG"></category></entry><entry><title>Blockchain-Based Carbon Trading for Stakeholders in Brazil</title><link href="https://watch.eeg.cl.cam.ac.uk/w/f1Uxw34FRLEfVNBBpzbsgD" rel="alternate"></link><published>2025-03-28T17:09:11Z</published><updated>2025-03-28T17:09:11Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/f1Uxw34FRLEfVNBBpzbsgD</id><summary type="html"><p>Full Title:<br />+Democratizing Carbon Markets: A Blockchain-Based Emission Trading System for Small and Large-Scale Stakeholders in Brazil</p>+The integration of blockchain technology into carbon markets offers a unique opportunity to create more transparent, inclusive, and efficient trading mechanisms. This presentation introduces a novel Blockchain Emission Trading System (BETS) model designed to align with Brazil’s new carbon market legislation (Law 15042/2024), ensuring that both large landholders and small rural producers can participate fairly. Our approach leverages official land registries, such as SICAR, to create spatially and temporally verifiable carbon credits, preventing fraud and double counting while enabling greater accessibility for smaller stakeholders who often struggle to enter regulated markets. By decentralizing the issuance and trading of carbon credits, our model aims to reduce intermediaries, lower costs, and promote broader participation, ultimately fostering a more equitable environmental and economic transition. Through a systematic mapping study, we identify key challenges and research directions for blockchain-based carbon markets and propose a framework that ensures compliance with national and international standards while prioritizing social and economic inclusivity.</p>+Jean is a professor at the Federal University of Santa Catarina (UFSC) in Brazil, specializing in information security, blockchain technology, and electronic documents. He holds a PhD in Computer Science from the University of Cambridge, where his research focused on cryptographic protocols and secure execution of code. Over the years, he has worked extensively on the development of blockchain-based solutions, particularly in the areas of digital identity, electronic signatures, and regulatory compliance. His recent work explores the use of blockchain to improve transparency, security, and inclusivity in digital ecosystems, including its application in carbon markets and sustainable finance.</p></summary><category term="EEG Seminars"></category></entry><entry><title>New preprint on benchmarking ultra-low power neural accelerators</title><link href="https://anil.recoil.org/news/2025-npu-bench-1" rel="alternate"></link><published>2025-03-28T00:00:00Z</published><updated>2025-03-28T00:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/news/2025-npu-bench-1</id><summary type="html"><p><a class="contact" href="https://profiles.imperial.ac.uk/joshua.millar22">Josh Millar</a> just released our latest preprint on how to make sense of the growing number of dedicated, ultra-low-power 'neural network accelerators' that are found in many modern embedded chipsets. My interest in this derives from wanting to decouple from the cloud when it comes to <a href="https://anil.recoil.org/projects/osmose">low-latency local environments</a>, and this needs fast tensor operations in hardware. Josh found a huge number of interesting NPUs in modern low-cost chips, ranging from <a href="https://www.espressif.com/en/products/socs/esp32">ESP32</a>-based boards over to <a href="https://arm.com">ARM</a> ones. All of these have quite a variety of tradeoffs, from the operations supported (which affects which models can be run on them) to the amount of memory and CPU power.</p>···</article></summary><category term="ryan.freumh.org"></category></entry><entry><title>2025-03-23 Weekly Notes</title><link href="https://ancazugo.github.io/research/outreach/2025/03/23/weekly-notes.html" rel="alternate"></link><published>2025-03-23T00:00:00Z</published><updated>2025-03-23T00:00:00Z</updated><author><name>AndrésZG</name></author><id>https://ancazugo.github.io/research/outreach/2025/03/23/weekly-notes</id><summary type="html"><p>After the issues and errors with generating the tree data for all of England, I realised that one of the best ways to proof-check the data was to visualize it and more importantly, to let others see it so they can check things that I might be missing. Because of this, I spent a good chunck of time trying to create a web map of all the trees in London. I used CARTO to do it but it wasn’t straightforward, mostly because of the size of the data. My files were stored in parquet format which apparently CARTO nor Mapbox are capable of transforming automatically to web Mercator, so I had to change the CRS in Sedona and re-generate the parquet partitions. Unfortunately, knowing where each partition is pointed at is not very easy, so what I did was geo-reference using the UK National Grid and then generate the partitions. The result is the map below where all the trees are shown with radius meaning the crown area and the colour the height. I want to improve it because at the moment the layers represent the partitions of the parquet files, but I know you can store the files in BigQuery or S3 and then use CARTO to visualize them (anyone willing to help, I’ll buy a beer and if the weather is good, an ice cream). Nonetheless, I will try that after the paper is done. Speaking of, I realised that I didn’t want to just present correlations between the tree data, remote sensing metrics and deprivation. That seems only descriptive statistics, but following the advice of <a href="https://anil.recoil.org/">Anil</a> from previous meetings, I’ve also worked on a metric that combines all of the variables into one. My original idea was to represent environmental deprivation, as the IMD does, but my variables don’t measure that intrinsecally, they just present where more nature is “available”, so I guess it will be more of a nature availability index. It relies on green and blue infrastructure and it is built on the weights from the first two components of a PCA for the entire country. The cool thing is that it can actually tell which geographic areas are urban or rural and you can see the deprivation pattern in the data. However, the logic behind them is just additive, as in it’s just following the form $y = a_1x_1 + … + a_ix_i$, where each $x$ is the “nature metric” for a given geographical area, and each coefficient $a$ is associated with said metric. This would be a-dimensional as variables were normalised. Then, after this I would build a Gini Coefficient of green disparity per Local Authority (this could be done at greater geographic scales such as region). This is based on this <a href="https://www.sciencedirect.com/science/article/pii/S0169204624002305#s0025">paper</a>, but also comments made by <a href="https://www.sustainabledesign.arct.cam.ac.uk/">Ronita</a> on how to represent my metrics, as the Gini Coefficient is widely used in economics when talking about wealth inequality.</p>-<p>Finally, on a lighter note, Saturday morning after a good tennis session, I went to the Computer Lab to attend a workshop on <a href="https://sonic-pi.net/">Sonic Pi by its creator Sam Aaron</a>, as part of the Cambridge Festival. It turned out, the attendants were just as old as the first movie in the Star Wars Sequel Trilogy (The Force Awakens in 2015). 😅 However, I ended up enjoying other activities during the festival, namely the Insomia AI project and the earphones that uses AI to monitor your heart rate. More importantly, expect my appearance in the CST social media channels as I answer questions on super computers, asked by none other than my dear friend and fellow PhD student <a href="https://www.onkargulati.com/">Onkar Gulati</a>.</p></summary><category term="Andrés Zúñiga-González’s Blog"></category></entry><entry><title>LIFE becomes an Official Statistic of the UK government</title><link href="https://anil.recoil.org/notes/life-official-statistic" rel="alternate"></link><published>2025-03-21T00:00:00Z</published><updated>2025-03-21T00:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/notes/life-official-statistic</id><summary type="html"><p>Our <a href="https://anil.recoil.org/papers/2024-life">recently published</a> <a href="https://anil.recoil.org/projects/life">LIFE</a> biodiversity metric has just been integrated into a newly recognised <a href="https://defraenvironment.blog.gov.uk/2025/01/20/newly-recognised-official-statistic-tracks-the-environmental-impact-of-our-consumption/">Official Statistic from the UK government</a>! This integrates the core LIFE biodiversity metric with <a href="https://anil.recoil.org/papers/2024-food-life">food provenance data</a> to track the environmental impacts of our consumption habits.</p>+<p>Finally, on a lighter note, Saturday morning after a good tennis session, I went to the Computer Lab to attend a workshop on <a href="https://sonic-pi.net/">Sonic Pi by its creator Sam Aaron</a>, as part of the Cambridge Festival. It turned out, the attendants were just as old as the first movie in the Star Wars Sequel Trilogy (The Force Awakens in 2015). 😅 However, I ended up enjoying other activities during the festival, namely the Insomia AI project and the earphones that uses AI to monitor your heart rate. More importantly, expect my appearance in the CST social media channels as I answer questions on super computers, asked by none other than my dear friend and fellow PhD student <a href="https://www.onkargulati.com/">Onkar Gulati</a>.</p></summary><category term="Andrés Zúñiga-González’s Blog"></category></entry><entry><title>Global Maps of Human Threats to Biodiversity and Species Extinctions</title><link href="https://watch.eeg.cl.cam.ac.uk/w/3exAV8tLbnPSGqoKv2mZts" rel="alternate"></link><published>2025-03-21T16:11:39Z</published><updated>2025-03-21T16:11:39Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/3exAV8tLbnPSGqoKv2mZts</id><summary type="html"><p>Full title:<br />+Towards Global Maps of Anthropogenic Threats to Biodiversity and Their Contributions to Species Extinctions</p>+Species extinctions are primarily driven by loss of habitat, which is relatively easy to monitor by satellite remote sensing; other anthropogenic threats to biodiversity, like hunting, are much more difficult to observe directly. My PhD project draws on local studies which capture the population effect of some anthropogenic threat, scaling these results using machine learning and remote sensing. In this talk, I will discuss my first attempt at this through quantifying species-specific responses to hunting pressure. I find that machine learning methods can offer marked improvements over (linear) statistical models, which are commonly used in ecology, but model validation must be done carefully to properly contextualise predictive performance. I will preview my plans for integrating these hunting pressure models with the LIFE biodiversity metric framework to express pressure in terms of extinction risk. If there is time, I will also discuss future plans for my PhD.</p>+Emilio is a PhD student in the Department of Zoology at the University of Cambridge in the Conservation Science Group and the Energy and Environment Group. He is supervised by Andrew Balmford, with co-supervision from Anil Madhavapeddy and Tom Swinfield. He is also part of the AI for Environmental Risks Centre for Doctoral Training, a researcher at the Cambridge Centre for Carbon Credits, and a member of Churchill College. His research focuses on the uses of predictive modeling for biodiversity conservation, with an emphasis on quantifying species-specific responses to human disturbance.</p></summary><category term="EEG Seminars"></category></entry><entry><title>LIFE becomes an Official Statistic of the UK government</title><link href="https://anil.recoil.org/notes/life-official-statistic" rel="alternate"></link><published>2025-03-21T00:00:00Z</published><updated>2025-03-21T00:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/notes/life-official-statistic</id><summary type="html"><p>Our <a href="https://anil.recoil.org/papers/2024-life">recently published</a> <a href="https://anil.recoil.org/projects/life">LIFE</a> biodiversity metric has just been integrated into a newly recognised <a href="https://defraenvironment.blog.gov.uk/2025/01/20/newly-recognised-official-statistic-tracks-the-environmental-impact-of-our-consumption/">Official Statistic from the UK government</a>! This integrates the core LIFE biodiversity metric with <a href="https://anil.recoil.org/papers/2024-food-life">food provenance data</a> to track the environmental impacts of our consumption habits.</p><p>I must admit that I'd not heard of &quot;Official Statistics&quot; before this, so I did a bit of research. The UK <a href="https://osr.statisticsauthority.gov.uk/">Office for Statistics Regulation</a> says that:</p><p>Official statistics are statistics produced by Crown bodies and other organisations listed within an Official Statistics Order, on behalf of the UK government or devolved administrations.···</article></summary><category term="ryan.freumh.org"></category></entry><entry><title>2025-03-16 Weekly Notes</title><link href="https://ancazugo.github.io/research/outreach/2025/03/16/weekly-notes.html" rel="alternate"></link><published>2025-03-16T00:00:00Z</published><updated>2025-03-16T00:00:00Z</updated><author><name>AndrésZG</name></author><id>https://ancazugo.github.io/research/outreach/2025/03/16/weekly-notes</id><summary type="html"><p>Most of my week was devoted to debugging my code to count the number of trees per geographical area. As mentioned in previous posts, I’ve been using Apache Sedona to process all my vector files, which has reduced the computation time significantly, particularly for spatial joins around all buildings in England. However, I realised that I hadn’t actually counted how many trees there are in total, which sounds trivial, but it turned out to be not as easy with the code that I had. So it turns out that when you are doing a Spatial Join with an Spatial RDD, you are only expected to pass two columns, the identifier and geometry for each row. This is not present in the docs, but I found it by accident while converting from Spark DataFrames to RDDs. Also, due to the number of trees (in the hundreds of millions range for England. Urban Cambridge apparently has ~66K 😉), I had to increase the memory requirements of the Sedona config but then it didn’t run, so what I ended up doing was iteratating over each Local Authority and querying the trees for that area, while counting the number of trees in each LSOA (Local Authorities are made by many LSOAs). This takes about 2 hours to run but was the best solution I found to overcome the Spark errors. Also, I realised that for ~300 LSOAs, I wasn’t getting any trees but it’s because their raster files, even though read by Rasterio, are corrupted due to inlavid values. It was very tricky to find the solution to this but a simple call to the object’s values can be caught by the <code class="language-plaintext highlighter-rouge">try/except</code> clause so that solved it for now. The good news is that all of these corrections are done and now I (hope to) finally have a clean and complete dataset for analysis which will be my main goal for this week, so I can complete the results and discussion of this paper. And right on time because it´s the end of the term so my focus will be this on for the start of the Easter break. Speaking of, at some point in the next weeks I will write an in-depth blog about working with Sedona after the paper is done, because I feel that a lot of people (academics in particular) would benefit from it since the tool is in its early adoption phase in the industry, while very few academics use it in their workflow.</p><p>On a side note, my supervisions for the academic year finished last week. I was teaching the Maths and Programming course for the new design tripos in the Department of Architecture. It was a very nurturing experience that allowed me to teach in the Cambridge system, which is very unique, but also gave me the chance to re-learn some of the mathematical concepts, notably linear algebra and differential calculus that are behind ML algorithms. I had learned those more than 10 years ago when I was in Uni, and I was surprised of how rusty I was with my math. Related to this, I just started reading <a href="https://www.penguin.co.uk/books/446849/why-machines-learn-by-ananthaswamy-anil/9780241586488">Why Machines Learn</a> by Anil Ananthaswamy, for a more in-depth refresher on Maths. Will come back to this with my comments once I finish it.-Finally, I successfully finished the intermediate 1 French course (sort of B1), which means that I can say <em>Je ne sais pas</em> when I don’t understand why my code behaves in a unexpected way. 🤣</p></summary><category term="Andrés Zúñiga-González’s Blog"></category></entry><entry><title>A fully AI-generated paper just passed peer review; notes from our evidence synthesis workshop</title><link href="https://anil.recoil.org/notes/ai-for-evidence-synthesis-workshop" rel="alternate"></link><published>2025-03-12T00:00:00Z</published><updated>2025-03-12T00:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/notes/ai-for-evidence-synthesis-workshop</id><summary type="html"><p>Access to reliable and timely scientific evidence is utterly vital for the practise of responsible policymaking, especially with all the turmoil in the world these days. At the same time, the evidence base on which use to make these decisions is rapidly morphing under our feet; the <a href="https://sakana.ai/ai-scientist-first-publication/">first entirely AI-generated paper passed peer review</a> at an ICLR workshop today. We held a workshop on this topic of AI and evidence synthesis at <a href="https://pem.cam.ac.uk">Pembroke College</a> last week, to understand both the opportunities for the use of AI here, the <a href="https://anil.recoil.org/papers/2024-ce-llm">strengths and limitations</a> of current tools, areas of progress and also just to chat with policymakers from <a href="https://www.gov.uk/government/organisations/department-for-science-innovation-and-technology">DSIT</a> and thinktanks about how to approach this rapidly moving area.</p>+Finally, I successfully finished the intermediate 1 French course (sort of B1), which means that I can say <em>Je ne sais pas</em> when I don’t understand why my code behaves in a unexpected way. 🤣</p></summary><category term="Andrés Zúñiga-González’s Blog"></category></entry><entry><title>Identifying Key Countries in the Illegal Elephant Ivory Trade Networ</title><link href="https://watch.eeg.cl.cam.ac.uk/w/oW6eqJBH1Hkwu6wE7XzQT3" rel="alternate"></link><published>2025-03-14T17:34:06Z</published><updated>2025-03-14T17:34:06Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/oW6eqJBH1Hkwu6wE7XzQT3</id><summary type="html"><p>Abstract:</p>+<p>Illegal wildlife trade is a key driver of biodiversity loss, but targeting policy to maximise disruption to trade remains a key challenge. A network approach was applied to seizure data to prioritise national action disrupting the illegal trade of elephant ivory. By simulating the removal of countries from trade, targeting groups of countries was found to be most effective due to network redundancy. Despite temporal variability, trade was highly concentrated and cessation in less than 10 countries would have disrupted 75% of trade in 2018-2020. These findings support evidence-based legislation and efficient allocation of conservation resources for tackling illegal wildlife trade.</p>+<p>Jakob is a PhD student in the Conservation and Development Lab (Department of Geography). His research focusses on evaluating policy for sustainable land systems, supervised by Prof. Rachael Garrett and Prof. Srinivasan Keshav. This work is supported by the Centre for Doctoral Training on Artificial Intelligence applied to the study of Environmental Risk (AI4ER CDT). Before starting his PhD, Jakob completed an MRes with AI4ER in Environmental Data Science, where he collaborated with TRAFFIC to develop data-driven tools to inform international illegal wildlife trade policy. Previously, Jakob completed an undergraduate degree in Natural Sciences at the University of Cambridge, specialising in Plant Sciences, and contributed to research on metrics for biodiversity offsetting, novel approaches to wildlife monitoring and forest ecology.</p></summary><category term="EEG Seminars"></category></entry><entry><title>A fully AI-generated paper just passed peer review; notes from our evidence synthesis workshop</title><link href="https://anil.recoil.org/notes/ai-for-evidence-synthesis-workshop" rel="alternate"></link><published>2025-03-12T00:00:00Z</published><updated>2025-03-12T00:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/notes/ai-for-evidence-synthesis-workshop</id><summary type="html"><p>Access to reliable and timely scientific evidence is utterly vital for the practise of responsible policymaking, especially with all the turmoil in the world these days. At the same time, the evidence base on which use to make these decisions is rapidly morphing under our feet; the <a href="https://sakana.ai/ai-scientist-first-publication/">first entirely AI-generated paper passed peer review</a> at an ICLR workshop today. We held a workshop on this topic of AI and evidence synthesis at <a href="https://pem.cam.ac.uk">Pembroke College</a> last week, to understand both the opportunities for the use of AI here, the <a href="https://anil.recoil.org/papers/2024-ce-llm">strengths and limitations</a> of current tools, areas of progress and also just to chat with policymakers from <a href="https://www.gov.uk/government/organisations/department-for-science-innovation-and-technology">DSIT</a> and thinktanks about how to approach this rapidly moving area.</p><p><em>(The following notes are adapted from jottings from <a class="contact" href="https://www.cst.cam.ac.uk/people/jkm40">Jessica Montgomery</a>,<a class="contact" href="https://samreynolds.org/">Sam Reynolds</a>, <a class="contact" href="https://ai.cam.ac.uk/people/annabelle-scott">Annabelle Scott</a> and myself. They are not at all complete, but hopefully useful!)</em></p><p>We invited a range of participants to the workshop and held it at Pembroke College (the choice of the centuries-old location felt appropriate).···<a href="https://anil.recoil.org/ideas/computational-storage-for-vector-dbs">usecases</a> for using heterogenous hardware like FPGAs these days.There's a great <a href="https://signalsandthreads.com/programmable-hardware/">Signals and Threads episode</a> on-programmable hardware with <a class="contact" href="https://github.com/andrewray">Andy Ray</a> that should give you more useful background knowledge as well.</p></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>2025-02-28 Weekly Notes</title><link href="https://ancazugo.github.io/research/outreach/2025/02/28/weekly-notes.html" rel="alternate"></link><published>2025-02-28T00:00:00Z</published><updated>2025-02-28T00:00:00Z</updated><author><name>AndrésZG</name></author><id>https://ancazugo.github.io/research/outreach/2025/02/28/weekly-notes</id><summary type="html"><p>This week I focused mostly on checking why I had some empty rows in the estimation of the canopy cover and the tree count from the Vegetation Object Model. After investigating the code, I found that the issue was behind a try/except clause that was masking corrupted files. Now, this was the intended behaviour originally, as some files can’t be read, however, because the data processing is done at the geographic statistical level (read <a href="https://www.ons.gov.uk/census/census2021dictionary/areatypedefinitions">ONS</a>), and the VOM tiles come in 5x5 km files (following GB National Grid), even when one file can’t be read, the other ones wouldn’t be processed either. With the new small correction, only the corrupted files are not processed and the rest of the files are processed as intended. So from the 5402 tiles, only 26 can’t be read.</p>+programmable hardware with <a class="contact" href="https://github.com/andrewray">Andy Ray</a> that should give you more useful background knowledge as well.</p></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>An Introductory Guide on How to Create Beautiful Slides</title><link href="https://watch.eeg.cl.cam.ac.uk/w/dFShkouits1FFyUctiSSH5" rel="alternate"></link><published>2025-02-28T17:59:20Z</published><updated>2025-02-28T17:59:20Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/dFShkouits1FFyUctiSSH5</id><summary type="html"><p>Frank Feng is a first-year Ph.D. student in the Department of Computer Science and Technology at the University of Cambridge. His research interests lie at the intersection of machine learning and earth sciences, with a particular focus on the application of self-supervised learning in remote sensing.</p></summary><category term="EEG Seminars"></category></entry><entry><title>2025-02-28 Weekly Notes</title><link href="https://ancazugo.github.io/research/outreach/2025/02/28/weekly-notes.html" rel="alternate"></link><published>2025-02-28T00:00:00Z</published><updated>2025-02-28T00:00:00Z</updated><author><name>AndrésZG</name></author><id>https://ancazugo.github.io/research/outreach/2025/02/28/weekly-notes</id><summary type="html"><p>This week I focused mostly on checking why I had some empty rows in the estimation of the canopy cover and the tree count from the Vegetation Object Model. After investigating the code, I found that the issue was behind a try/except clause that was masking corrupted files. Now, this was the intended behaviour originally, as some files can’t be read, however, because the data processing is done at the geographic statistical level (read <a href="https://www.ons.gov.uk/census/census2021dictionary/areatypedefinitions">ONS</a>), and the VOM tiles come in 5x5 km files (following GB National Grid), even when one file can’t be read, the other ones wouldn’t be processed either. With the new small correction, only the corrupted files are not processed and the rest of the files are processed as intended. So from the 5402 tiles, only 26 can’t be read.</p><p>In addition to this, I wanted to reduce the over segmentation of trees from the VOM, because I was getting way more tree points than there actually are (at least visually), which leads to an overestimation of the number of trees. To do so, two changes were made thanks to chatGPT 🙃:</p>···<p>We had a meeting this week about future LIFE metric work, and so I need to write up bits of that, as we did sensibly set a deadline for some progress updates in a month!</p>-<p>I still owe Simon Tarr a way to run the AoH pipeline via Docker that I said I'd do a few weeks ago, and I need to follow up with Chess on our differences in species filtering - though hopefully an easier thing to track with my new reporting output.</p></summary><category term="Digital Flapjack"></category></entry><entry><title>2025-02-21 Weekly Notes</title><link href="https://ancazugo.github.io/research/outreach/2025/02/21/weekly-notes.html" rel="alternate"></link><published>2025-02-21T00:00:00Z</published><updated>2025-02-21T00:00:00Z</updated><author><name>AndrésZG</name></author><id>https://ancazugo.github.io/research/outreach/2025/02/21/weekly-notes</id><summary type="html"><p>So back to the routine after coming from Italy for BIOSPACE25. It was a really great experience in the ESA HQ but I got a cold from Italy 😷.</p>+<p>I still owe Simon Tarr a way to run the AoH pipeline via Docker that I said I'd do a few weeks ago, and I need to follow up with Chess on our differences in species filtering - though hopefully an easier thing to track with my new reporting output.</p></summary><category term="Digital Flapjack"></category></entry><entry><title>Robotics and Sensing for Sustainable Crop Production</title><link href="https://watch.eeg.cl.cam.ac.uk/w/9hADtA5Fov2vdDt9iNVjJQ" rel="alternate"></link><published>2025-02-21T16:26:06Z</published><updated>2025-02-21T16:26:06Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/9hADtA5Fov2vdDt9iNVjJQ</id><summary type="html"><p>Abstract:<br />+Crop farming is essential in our society, providing food, feed, fiber, and fuel. We heavily rely on crop production, but at the same time, we need to reduce the production footprint. We aim to address this key challenge by investigating new solutions to produce crops more sustainably. We study novel technology-driven approaches to move toward sustainable crop production. Agricultural robots offer promising directions to address management challenges in agricultural fields or support plant breeding efforts through large-scale trait acquisition. For that, field robots need the ability to perceive and model their environment, predict possible future developments, and make appropriate decisions in complex and changing situations. This talk will showcase our recent developments in robotics for crop production, incorporating machine learning to support farmers in operating more sustainably and reducing some negative impacts on the ecosystem.</p>+Cyrill Stachniss is a full professor at the University of Bonn and heads the Photogrammetry and Robotics Lab. He is also a Visiting Professor in Engineering at the University of Oxford and is with the Lamarr Institute for Machine Learning and Artificial Intelligence. Before his appointment in Bonn, he was with the University of Freiburg and ETH Zurich. Since 2010, he has been a Microsoft Research Faculty Fellow and received the IEEE RAS Early Career Award in 2013. From 2015 to 2019, he was senior editor for the IEEE Robotics and Automation Letters. He is the spokesperson of the DFG Cluster of Excellence "PhenoRob" at the University of Bonn, together with his colleague Heiner Kuhlmann. His research focuses on probabilistic techniques as well as learning approaches for mobile robotics, perception, and navigation. The main application areas of his research are autonomous service robots, agricultural robotics, and self-driving cars. He has co-authored over 300 publications and has coordinated multiple large-scale research projects on the national and European levels. Besides his university involvement, he cofounded three startups: Escarda Technologies, DeepUp, and PhenoInspect.</p></summary><category term="EEG Seminars"></category></entry><entry><title>2025-02-21 Weekly Notes</title><link href="https://ancazugo.github.io/research/outreach/2025/02/21/weekly-notes.html" rel="alternate"></link><published>2025-02-21T00:00:00Z</published><updated>2025-02-21T00:00:00Z</updated><author><name>AndrésZG</name></author><id>https://ancazugo.github.io/research/outreach/2025/02/21/weekly-notes</id><summary type="html"><p>So back to the routine after coming from Italy for BIOSPACE25. It was a really great experience in the ESA HQ but I got a cold from Italy 😷.</p><p>This week started by going to the Social and Spatial Epidemiology Module from the <a href="https://www.phs.group.cam.ac.uk/">PHS MPhil</a> in the Biomedical Campus. It’s originally desgined to help public health students understand that it is a spatial domain, with all that it means. I really enjoyed the course in that I was exposed to things that I wasn’t aware of like redlining and gerrymandering in the US.</p>···<p>This largely involved talking to individual publishers and agreeing not to directly train generative AI models and to keep them private to our own research use. Fairly reasonable stuff.</p>-<span><a class="fn-label" href="https://anil.recoil.org/#ref-1-fn-1">↩︎︎</a></span></li></ol></section></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>BIOSPACE25 Blog</title><link href="https://ancazugo.github.io/research/outreach/2025/02/14/biospace25-blog.html" rel="alternate"></link><published>2025-02-14T00:00:00Z</published><updated>2025-02-14T00:00:00Z</updated><author><name>AndrésZG</name></author><id>https://ancazugo.github.io/research/outreach/2025/02/14/biospace25-blog</id><summary type="html"><h2 id="day-2">Day 2</h2>+<span><a class="fn-label" href="https://anil.recoil.org/#ref-1-fn-1">↩︎︎</a></span></li></ol></section></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Using Low-cost, Research-led, Decentralised Networks</title><link href="https://watch.eeg.cl.cam.ac.uk/w/8PhivRm85jZuFg8v55yo7F" rel="alternate"></link><published>2025-02-14T16:25:58Z</published><updated>2025-02-14T16:25:58Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/8PhivRm85jZuFg8v55yo7F</id><summary type="html"><p>Full Title:<br />+Using Low-cost, Research-led, Decentralised Networks to Increase Access to High Quality Microspatial Data on Building Stocks, and the Built and Natural Infrastructure</p>+The Colouring Cities Research Programme (CCRP) is overseen by an informal, international academic consortium that uses its decentralized research-led network, to co-create and manage, permanent open data/visualisation platforms across countries. These provide standardised, open microspatial data on the characteristics, performance, and short/long-term dynamics of building stocks, and built and natural infrastructure. They also test feedback loops between live streaming, computational inference, and crowdsourcing approaches to improve coverage and reliability of data, and to support cross sector/multidisciplinary engagement. Research Institutions from 30 countries are currently involved. The programme has been set up to accelerate progress towards UN SDGs; to provide access to big data required to exploit the potential of AI &amp; ML and gain insights at scale; to reduce research costs and overlaps and speed up testing of research applications by pooling expertise, funding and ideas, and; to ensure that areas such as data standardisation, uncertainty in data, and citizen privacy and security are prioritised, as commercial demand for microspatial data grows.</p>+Polly Hudson is a Senior Research Fellow at The Alan Turing Institute and PI for the Colouring Cities Research Programme. She was previously a Senior Research Fellow at the Centre for Advanced Spatial Analysis University College and held a Visiting Fellowship at the Kellogg Centre for the Historic Environment, University of Oxford. Relevant appointments include advisory/board positions for the Department of Culture, Media and Sport, English Heritage, The Royal Institute of British Architects, and the National Lottery (charitable arm).</p>+<p>Polly trained as an architectural historian, and initially worked in furniture making, historic building restoration, museum design, and community planning. In 1996 she set up the Building Exploratory charitable trust in London as a prototype for multidisciplinary knowledge sharing centres about local building stocks, which was co-built over 6 years by citizens, local and central government, industry, non-profits &amp; academia. The first iteration of the Colouring Cities mapping platform interface was tested in 1998. Between 2014 and 2019 Polly received EPSRC funding from UCL to develop the CCRP concept within academia where she partnered with Tom Russell who built the back and front end for the London prototype &amp; advised on open licences. Since 2020 her position as CCRP PI has been funded by The Alan Turing Institute.</p></summary><category term="EEG Seminars"></category></entry><entry><title>BIOSPACE25 Blog</title><link href="https://ancazugo.github.io/research/outreach/2025/02/14/biospace25-blog.html" rel="alternate"></link><published>2025-02-14T00:00:00Z</published><updated>2025-02-14T00:00:00Z</updated><author><name>AndrésZG</name></author><id>https://ancazugo.github.io/research/outreach/2025/02/14/biospace25-blog</id><summary type="html"><h2 id="day-2">Day 2</h2><h3 id="my-presentation-available-to-watch-here">My presentation (available to watch <a href="https://www.youtube.com/live/e-eQ8XhRrsE?si=lv0p4CV3kdKGxmTk&amp;t=14326">here</a>)</h3>···-</article></summary><category term="ryan.freumh.org"></category></entry><entry><title>The Partially Dynamic Web</title><link href="https://digitalflapjack.com/blog/the-partially-dynamic-web/" rel="alternate"></link><published>2024-12-08T11:59:48Z</published><updated>2024-12-08T11:59:48Z</updated><author><name>MichaelD</name></author><id>https://digitalflapjack.com/blog/the-partially-dynamic-web/</id><summary type="html"><h1>Background</h1>+</article></summary><category term="ryan.freumh.org"></category></entry><entry><title>Optimising Sustainable Energy with Functional Programming</title><link href="https://watch.eeg.cl.cam.ac.uk/w/feDup1JutmgQkC6ipGF9r5" rel="alternate"></link><published>2024-12-13T18:23:12Z</published><updated>2024-12-13T18:23:12Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/feDup1JutmgQkC6ipGF9r5</id><summary type="html"><p>Abstract:</p>+<p>This talk describes some results from a collaboration between Computer Science, Physics, and Climate Impact Research on theories and tools for performance optimisation of strongly coupled physical systems with a large parameter space. The first part of the talk discusses computing optimal policies; we have used these techniques for climate decisions and for fusion energy designs. The second part of the talk will focus on one particularly important concept: the Pareto-front, which mathematically captures the trade-offs between two (or more) conflicting objectives. The core object of study is an expensive black-box function computing multiple objectives, for which we approximate the Pareto front using adaptive mesh refinement.</p>+<p>Patrik Jansson is a professor in the Computer Science and Engineering Department, joint between Chalmers University of Technology and the University of Gothenburg, Sweden. His main research areas are Programming Languages, Functional Programming, Domain-Specific Languages, and their application to climate, physics, etc. His research focus is on systems for constructing correct and reusable software. The goal is to develop the programming languages of the future and theories, tests and proofs of the correctness of high-level models of complex systems. Important techniques include functional programming, domain-specific languages and type theory. Examples of applications are climate impact research, physics, and language technology but many results are also curiosity-driven basic research with generic applicability in most areas.</p>+<p>Patrik has been on sabbatical in Oxford, as a Visiting Fellow of Kellogg College for Michaelmas term 2024, visiting Prof Jeremy Gibbons.</p></summary><category term="EEG Seminars"></category></entry><entry><title>The Partially Dynamic Web</title><link href="https://digitalflapjack.com/blog/the-partially-dynamic-web/" rel="alternate"></link><published>2024-12-08T11:59:48Z</published><updated>2024-12-08T11:59:48Z</updated><author><name>MichaelD</name></author><id>https://digitalflapjack.com/blog/the-partially-dynamic-web/</id><summary type="html"><h1>Background</h1><p>I have three websites (this one, my <a href="https://mynameismwd.org/">personal site</a>, and one for <a href="https://mwdales-guitars.uk/">my luthiery endeavours</a>), and despite each starting out with a different technology stack, for the last few years I’d migrated them all to the <a href="https://gohugo.io">Hugo static site generator</a>, as a way of making it easier for me to mess around with. Without a fixed database, I could more easily readily structure the content as I wanted it, I had more freedom over templating, and ultimately it’s less resource intensive to compile the site occasionally and just serve static files than keep dynamic infrastructure running for what is a set of low traffic websites. At least in theory, we’ll come back to this last point.</p><p>Like most static site generators, Hugo uses a system called <a href="https://gohugo.io/content-management/front-matter/">Front matter</a>, where you store each page as a <a href="https://en.wikipedia.org/wiki/Markdown">markdown file</a> of content, and some <a href="https://en.wikipedia.org/wiki/YAML">YAML</a> at the top of that file to store metadata, such as page title and publication date, which isn’t something markdown supports. With these two parts Hugo can generate your website based on where files are in directories, and with the appropriate bits from the front matter. Hugo will use a template system to turn your markdown into HTML files, roughly following the structure of the folders you store the markdown files in.</p><p>The templates for my sites I’d made by hand myself, which I think is a key part to unlocking the power of Hugo. Because not only can you decide how your markdown looks from your template, but you can also query the front matter, and so change how your page looks based on the metadata. I heavily used this feature, using it a bit like a database entry for each page. This let me add a synopsis to each page, or a title image with alt-text that becomes the thumbnail on the list views. For photos I store all the EXIF data in there too.</p>···<p>I’m using the <a href="https://aantron.github.io/dream/">Dream</a> library for <a href="https//ocaml.org/">OCaml</a>, which has both built in routing and templating. I made sure to keep the URL layout that Hugo used as best I could, so in theory the transition shouldn’t be noticed by most people, as all the content remains at the same URL it was.</p><p>Using a functional language for this kind of work actually maps very nicely: all I’m doing is taking data in one format and presenting it in another, so functional transforms are what I need. The way the website is stored on disk for a static site generator means I'm mostly doing a translation of that structure into the URIs for the website, so I was starting from a good place for this project.</p><p>Thanks to Hugo encouraging me to use shortcodes for all resources in a page (I never used the markdown image tags), it was low effort to ensure all resources in a page have their own URL to render them on demand, as I don’t ever need to parse the markdown myself beyond pulling out shortcodes. For images I’m just using <a href="https://gallium.inria.fr/camlimages/">Camlimages</a> which is quite an old library and doesn’t support all the image formats I have acquired over 20 plus years, but it’s enough to get started with. Performance wise, this will be a regression, as images are resized and stored in a small cache the first time they’re viewed, but given most people consume my site via RSS, when I add a new page and look at it myself to check it works, it’ll mean for most folk they don’t see that.</p>-<p>The aim so far has just been to get as close to the Hugo version as I can without changing the data on disk. What I'm looking forward to doing now I've switched is making changes to the on disk representation to let me simplify the OCaml code, and add some new fun features.</p></summary><category term="Digital Flapjack"></category></entry><entry><title>Horizon scan on AI and conservation published</title><link href="https://anil.recoil.org/news/2024-ai-conhorizon-1" rel="alternate"></link><published>2024-12-05T00:00:00Z</published><updated>2024-12-05T00:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/news/2024-ai-conhorizon-1</id><summary type="html"><p>Back in July 2024, a large group of conservation and computer scientists got together in the <a href="https://conservation.cam.ac.uk">CCI</a> to prioritise the storm of AI-related projects that have been kicking off around the world. Our key goal was to harness AI to accelerate the positive impact of conservation efforts, while minimising harm caused through either the direct or indirect use of AI technologies.</p>+<p>The aim so far has just been to get as close to the Hugo version as I can without changing the data on disk. What I'm looking forward to doing now I've switched is making changes to the on disk representation to let me simplify the OCaml code, and add some new fun features.</p></summary><category term="Digital Flapjack"></category></entry><entry><title>Smart Grid Trust Assessment</title><link href="https://watch.eeg.cl.cam.ac.uk/w/uFyApvuvALLv66D7x36FEr" rel="alternate"></link><published>2024-12-06T16:40:43Z</published><updated>2024-12-06T16:40:43Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/uFyApvuvALLv66D7x36FEr</id><summary type="html"><p>Abstract:</p>+<p>Energy systems are highly complex. State determination and detection of anomalies, faults or even attacks are only possible to a limited extent with traditional approaches. This talk will investigate how such systems can be planned and operated in the future in the area of conflict between high automation and trust by human operators.</p>+<p>Sebastian Lehnhoff is a Full Professor of Energy Informatics at the University of Oldenburg. He received his doctorate at the TU Dortmund University in 2009. Prof. Lehnhoff is chairman of the board of the OFFIS Institute for Information Technology and speaker of its Energy R&amp;D division. He is a board member of the section „Energy Informatics“ within the German Informatics Society (GI) as well as an active member of numerous committees and working groups focusing on ICT in future Smart Grids. In 2022 he was appointed to the Board of Trustees of the Volkswagen Foundation (VolkswagenStiftung). He is the CTO of openKONSEQUENZ e.G. – a registered cooperative industry association for the development of modular Open-Source SCADA/EMS. He serves as Chairman of the Executive Board of the Energy Research Centre of Lower Saxony (EFZN) as well as an Executive Committee Member of the ACM Special Interest Group on Energy Systems and Informatics (SIGEnergy). Prof. Lehnhoff is a member of the German Academy of Science and Engineering (acatech) as well as a member of the Berlin-Brandenburg Academy of Sciences and Humanities (BBAW).</p></summary><category term="EEG Seminars"></category></entry><entry><title>Horizon scan on AI and conservation published</title><link href="https://anil.recoil.org/news/2024-ai-conhorizon-1" rel="alternate"></link><published>2024-12-05T00:00:00Z</published><updated>2024-12-05T00:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/news/2024-ai-conhorizon-1</id><summary type="html"><p>Back in July 2024, a large group of conservation and computer scientists got together in the <a href="https://conservation.cam.ac.uk">CCI</a> to prioritise the storm of AI-related projects that have been kicking off around the world. Our key goal was to harness AI to accelerate the positive impact of conservation efforts, while minimising harm caused through either the direct or indirect use of AI technologies.</p><p>The first horizon scan resulting from this has just been published in Trends in Ecology and Evolution. If you're looking for a gentle introduction to some of the terms in AI from a non-experts perspective, the first section does a good job of defining a glossary as well. <span class="morewords">[…<a href="https://anil.recoil.org/news/2024-ai-conhorizon-1">118 words</a>]</span></p>···&nbsp; <a href="https://anil.recoil.org/papers/2024-loco-terracorder.pdf"><span class="nobreak">PDF<img alt="pdf" class="inline-icon" src="https://anil.recoil.org/assets/pdf.svg" /></span></a>-</blockquote></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Mon 18 Nov 2024</title><link href="https://ryan.freumh.org/2024-11-18.html" rel="alternate"></link><published>2024-11-18T00:00:00Z</published><updated>2024-11-18T00:00:00Z</updated><author><name>RyanG</name></author><id>https://ryan.freumh.org/2024-11-18.html</id><summary type="html"><article>+</blockquote></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Conservation Evidence</title><link href="https://watch.eeg.cl.cam.ac.uk/w/ijC1E36q7fn2qwxs7opSJq" rel="alternate"></link><published>2024-11-29T17:39:54Z</published><updated>2024-11-29T17:39:54Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/ijC1E36q7fn2qwxs7opSJq</id><summary type="html"><p>Grey literature’s inherent nature means that it is a difficult form of media to discover, typically being hidden deep within websites, analyse, following no standard file formats or structures, and process, due to the sheer volume of existing and actively produced literature, this forms a massive cost and time problem for organisations that require such literature in their function.<br />+We devise and implement a pipeline that uses Common Crawl internet archives to locate &amp; scrape potential grey literature; then process it for use in a multistage machine learning pipeline to classify and output relevant media.</p>+<p>Shrey Biswas is a second-year Computer Science Student at Pembroke College.<br />+Radhika Iyer is a second-year Computer Science Student at Murray Edwards College.<br />+Kacper Michalik is a Second-year Computer Science Student at Pembroke College</p></summary><category term="EEG Seminars"></category></entry><entry><title>Mon 18 Nov 2024</title><link href="https://ryan.freumh.org/2024-11-18.html" rel="alternate"></link><published>2024-11-18T00:00:00Z</published><updated>2024-11-18T00:00:00Z</updated><author><name>RyanG</name></author><id>https://ryan.freumh.org/2024-11-18.html</id><summary type="html"><article><span> Previous: <a href="2024-10-14.html">Mon 14 Oct 2024</a> </span>···<p>These conclude my live notes! Beyond the notes here, the corridor conversations were incredibly-useful for you to have put them online as well.</p></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Published a legal perspective on high integrity forest carbon credits</title><link href="https://anil.recoil.org/news/2024-cclr-carbon-1" rel="alternate"></link><published>2024-11-01T00:00:00Z</published><updated>2024-11-01T00:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/news/2024-cclr-carbon-1</id><summary type="html"><p><a class="contact" href="https://www.cst.cam.ac.uk/people/smc70">Sophie Chapman</a> lead an <a href="https://anil.recoil.org/ideas/legal-aspects-of-credits">effort</a> to explore a novel legal framework for forest carbon credits+useful for you to have put them online as well.</p></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Long-term Biodiversity Monitoring at Scale</title><link href="https://watch.eeg.cl.cam.ac.uk/w/pMzCFQKTrRtQ6jotF1z12V" rel="alternate"></link><published>2024-11-15T16:16:36Z</published><updated>2024-11-15T16:16:36Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/pMzCFQKTrRtQ6jotF1z12V</id><summary type="html"><p>Abstract:<br />+Comprehensive data on global biodiversity patterns is only obtainable through in-situ distributed sensor networks. However, these multi-device networks are constrained by battery lifetimes, must gather rich data from power-hungry sensors, and yet must be deployed in remote environments for long periods. We look at the feasibility of a prototype multi-sensor device using on-device reinforcement learning for power management.</p>+</ul></summary><category term="EEG Seminars"></category></entry><entry><title>A CarbonFirst Approach for Decarbonizing Cloud Computing</title><link href="https://watch.eeg.cl.cam.ac.uk/w/pxkLZ4jgVJMqjwZuhWicrK" rel="alternate"></link><published>2024-11-01T15:46:48Z</published><updated>2024-11-01T15:46:48Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/pxkLZ4jgVJMqjwZuhWicrK</id><summary type="html"><p>Abstract:</p>+<p>The exponential growth of cloud computing has been a defining trend of our time, fueled by rapidly growing demands from data-intensive and machine learning workloads. Despite the end of Dennard scaling, the cloud's energy demand grew more slowly than expected over the past decade due to the aggressive implementation of energy-efficiency optimizations. Unfortunately, there are few significant remaining optimization opportunities using traditional methods, and moving forward, the cloud's continued exponential growth will translate into rising energy demand, which, if left unchecked, will translate to increasing carbon emissions.</p>+<p>In this talk, I will argue for a CarbonFirst approach to designing cloud computing systems by making carbon efficiency a first-class design metric, similar to traditional metrics of performance and reliability. I will explain how today's systems can be made first carbon-aware by exposing energy and carbon usage information to software platforms and then made carbon-efficient by providing control over the system's carbon usage. I will present an initial design of a system to enable such carbon awareness and management and present several application case studies on how modern cloud applications can employ these mechanisms to reduce their carbon footprint. I will end with open research challenges in the emerging field of computational decarbonization.</p>+<p>Prashant Shenoy is currently a Distinguished Professor and Associate Dean in the College of Information and Computer Sciences at the University of Massachusetts Amherst. He received the B.Tech degree in Computer Science and Engineering from the Indian Institute of Technology, Bombay and the M.S and Ph.D degrees in Computer Science from the University of Texas, Austin. His research interests lie in distributed systems and networking, with a recent emphasis on cloud and sustainable computing. He has been the recipient of several best paper awards at leading conferences, including a Sigmetrics Test of Time Award. He is a fellow of the ACM, the IEEE, and the AAAS.</p></summary><category term="EEG Seminars"></category></entry><entry><title>Published a legal perspective on high integrity forest carbon credits</title><link href="https://anil.recoil.org/news/2024-cclr-carbon-1" rel="alternate"></link><published>2024-11-01T00:00:00Z</published><updated>2024-11-01T00:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/news/2024-cclr-carbon-1</id><summary type="html"><p><a class="contact" href="https://www.cst.cam.ac.uk/people/smc70">Sophie Chapman</a> lead an <a href="https://anil.recoil.org/ideas/legal-aspects-of-credits">effort</a> to explore a novel legal framework for forest carbon credits···<p>It is however, increadably slow - my graphics library isn't very well optimised, as usually on modern hardware it doesn't need it to push old-school VGA like graphics around, but running natively on my AMD Razen machine it was really quite poor, low single-digit frames per second. In part this I assume is related to Haiku not knowing about my fancy NVIDIA graphics card, and just using the stock framebuffer driver, and in part because OCaml doesn't know about Haiku enough to build a native binary and is instead using the bytecode backend.</p></summary><category term="Digital Flapjack"></category></entry><entry><title>Mapping greener futures with planetary computing</title><link href="https://anil.recoil.org/news/a0280750-2ef0-4f5c-b138-68f7b11b4c29-1" rel="alternate"></link><published>2024-10-24T01:00:00Z</published><updated>2024-10-24T01:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/news/a0280750-2ef0-4f5c-b138-68f7b11b4c29-1</id><summary type="html"><p>I got invited by <a href="https://profiles.ucl.ac.uk/78591-serta%C3%A7-sehlikoglu">Sertaç Sehlikoglu</a> to deliver a lecture to the Masters students down at the <a href="https://www.ucl.ac.uk/bartlett/igp/">UCL Institute for Global Prosperity</a>. I talked about the recent work on <a href="https://anil.recoil.org/projects/plancomp">planetary computing</a>, with an overview of the <a href="https://anil.recoil.org/projects/life">LIFE</a> and <a href="https://anil.recoil.org/papers/2024-food-life">FOOD</a> papers.</p>-<p><div class="video-center"></div></p></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Mon 14 Oct 2024</title><link href="https://ryan.freumh.org/2024-10-14.html" rel="alternate"></link><published>2024-10-14T01:00:00Z</published><updated>2024-10-14T01:00:00Z</updated><author><name>RyanG</name></author><id>https://ryan.freumh.org/2024-10-14.html</id><summary type="html"><article>+<p><div class="video-center"></div></p></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Challenges in Cyber-Physical Energy Systems - of Agents and Data</title><link href="https://watch.eeg.cl.cam.ac.uk/w/7XijwZ8ZtKtnLgRXKgt9G1" rel="alternate"></link><published>2024-10-18T16:03:59Z</published><updated>2024-10-18T16:03:59Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/7XijwZ8ZtKtnLgRXKgt9G1</id><summary type="html"><p>Abstract:</p>+<p>In this talk, first, some special challenges in cyber-physical energy systems will be reflected. Then, examples from research projects and field tests will be discussed to show how multi-agent systems can be used to tackle these challenges. Finally, the topic of research data management and its role in open research will be discussed.</p>+<p>Prof. Dr.-Ing. Astrid Nieße has been Professor for Digitalized Energy Systems at the University of Oldenburg since 2020 and a member of the Energy Division Board of the OFFIS - Institute of Computer Science. From 2018 to 2020 she was Professor for Energy Informatics at Leibniz University Hannover.</p>+<p>Astrid Nieße received her doctorate from the University of Oldenburg in 2015; her doctoral thesis dealt with the application of distributed algorithms in the field of decentralized energy systems .<br />+Astrid Nieße studied computer science and biology at the University of Bremen and at the University of Oldenburg.</p></summary><category term="EEG Seminars"></category></entry><entry><title>Mon 14 Oct 2024</title><link href="https://ryan.freumh.org/2024-10-14.html" rel="alternate"></link><published>2024-10-14T01:00:00Z</published><updated>2024-10-14T01:00:00Z</updated><author><name>RyanG</name></author><id>https://ryan.freumh.org/2024-10-14.html</id><summary type="html"><article><span> Previous: <a href="2024-10-07.html">Mon 7 Oct 2024</a> </span><span> Next: <a href="2024-11-18.html">Mon 18 Nov 2024</a> </span>···-</article></summary><category term="ryan.freumh.org"></category></entry><entry><title>On the SOCC 2024 PC</title><link href="https://anil.recoil.org/notes/socc-pc" rel="alternate"></link><published>2024-10-08T01:00:00Z</published><updated>2024-10-08T01:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/notes/socc-pc</id><summary type="html"><p>After some time away from cloud computing (due to my new focus on <a href="https://anil.recoil.org/projects/life">conservation research</a>), I served on the <a href="https://acmsocc.org/2024/">ACM SOCC 2024</a> program committee. It was quite interesting seeing the massive shift away from &quot;traditional&quot; cloud research (such as consensus protocols) towards many submissions aimed at accelerating machine learning workloads.</p>+</article></summary><category term="ryan.freumh.org"></category></entry><entry><title>Visual Digital Twins of Forests</title><link href="https://watch.eeg.cl.cam.ac.uk/w/9CqWsuQQykVtbuDPwuLwZs" rel="alternate"></link><published>2024-10-11T20:45:49Z</published><updated>2024-10-11T20:45:49Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/9CqWsuQQykVtbuDPwuLwZs</id><summary type="html"><p>Abstract:</p>+<p>Methods developed by the computer graphics community allow for the photorealistic rendering of complex geometry. In this talk we explore how such mathematical procedures can be leveraged to describe the growth, biomechanics, and combustion of trees at a detailed spatial level. These models facilitate a realistic 3D visualization of these processes at forest scale which allows exploring illustratively a variety of hypothetical environmental scenarios. Potential applications of such methods include the educational dissemination of environmental concepts, the generation of synthetic image data for training vision-based AI models, and the evaluation of ecological hypotheses expressed at plant organ scale.</p>+<p>Wojtek Palubicki is a Professor at Adam Mickiewicz University where he leads the Natural Phenomena Modelling Group. The research group uses methods from computer graphics and AI to describe and investigate natural pattern genesis. Before that he held a post-doctoral research scientist post at the SLCU developing mathematical models of plant developmental biology.</p></summary><category term="EEG Seminars"></category></entry><entry><title>On the SOCC 2024 PC</title><link href="https://anil.recoil.org/notes/socc-pc" rel="alternate"></link><published>2024-10-08T01:00:00Z</published><updated>2024-10-08T01:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/notes/socc-pc</id><summary type="html"><p>After some time away from cloud computing (due to my new focus on <a href="https://anil.recoil.org/projects/life">conservation research</a>), I served on the <a href="https://acmsocc.org/2024/">ACM SOCC 2024</a> program committee. It was quite interesting seeing the massive shift away from &quot;traditional&quot; cloud research (such as consensus protocols) towards many submissions aimed at accelerating machine learning workloads.</p><p>I also had a paper accepted there on <a href="https://anil.recoil.org/papers/2024-socc-murmuration">decentralised scheduling</a>, thanks to my former PhD student <a class="contact" href="https://www.cl.cam.ac.uk/~sv440/">Smita Vijayakumar</a> and her hard work on Murmuration!</p></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Mon 7 Oct 2024</title><link href="https://ryan.freumh.org/2024-10-07.html" rel="alternate"></link><published>2024-10-07T01:00:00Z</published><updated>2024-10-07T01:00:00Z</updated><author><name>RyanG</name></author><id>https://ryan.freumh.org/2024-10-07.html</id><summary type="html"><article><span> Previous: <a href="2024-09-30.html">Mon 30 Sep 2024</a> </span>···-</article></summary><category term="ryan.freumh.org"></category></entry><entry><title>Royal Society meeting on ecological/commercial risks</title><link href="https://anil.recoil.org/notes/rs-ecorisk-day1" rel="alternate"></link><published>2024-10-03T01:00:00Z</published><updated>2024-10-03T01:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/notes/rs-ecorisk-day1</id><summary type="html"><p>I'm at the Royal Society this morning for the 2 day programme on <a href="https://royalsociety.org/science-events-and-lectures/2024/10/ecological-and-commercial-risk/">&quot;How does ecological risk related to commercial risk?&quot;</a>, and am reporting on the <a href="https://royalsociety.org/-/media/events/2024/10/ecological-risk/programme-booklet.pdf">morning session</a>. The full program is being <a href="https://www.youtube.com/watch?v=gVuxzand8RE">livestreamed</a> so please do dial in if the below notes seem interesting to you. I put this note up almost live, so any errors below are my own.+</article></summary><category term="ryan.freumh.org"></category></entry><entry><title>Partner-driven Environmental Sensing:</title><link href="https://watch.eeg.cl.cam.ac.uk/w/pQBnfPWJi9kxLdeHY9YAA7" rel="alternate"></link><published>2024-10-04T19:15:38Z</published><updated>2024-10-04T19:15:38Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/pQBnfPWJi9kxLdeHY9YAA7</id><summary type="html"><p>Full Title: Partner-driven Environmental Sensing: Co-design with Indigenous Ojibwe Scientists and Malagasy Conservationists</p>+<p>Evolving environmental sensing technologies present a myriad of opportunities for gathering data to understand and promote environmental justice, biodiversity, and climate change mitigation. However, technical development from academic and commercial settings often struggle to translate to accessible solutions for marginalized communities. In this talk, I will explore the opportunities of partner-driven co-design, share the findings from a qualitative study of field scientists’ use of technology, and present two case studies: (1) designing environmental sensors with Indigenous Ojibwe scientists for manoomin (wild rice) conservation and (2) partnering with Malagasy conservation organizations to understand the role that technology can play in reforestation and biodiversity monitoring.</p>+<p>Eric Greenlee (he/him) is a PhD student in the College of Computing at Georgia Tech, co-advised by Ellen Zegura and Josiah Hester. Conducting research at the intersection of the Computing and Society Lab and the Ka Moamoa Lab, Eric explores partner-driven processes with communities often cut out of technology development to co-create emergent environmental sensors to address challenges in environmental justice, biodiversity loss, and climate change mitigation. By leveraging qualitative methods, he aims to strengthen connections across traditional silos to design and deploy user-friendly, networked, and low-power embedded systems. Prior to pursuing his PhD, Eric worked as a Radio Frequency engineer for the U.S. Federal Government and studied electrical engineering at Dartmouth College.</p></summary><category term="EEG Seminars"></category></entry><entry><title>Royal Society meeting on ecological/commercial risks</title><link href="https://anil.recoil.org/notes/rs-ecorisk-day1" rel="alternate"></link><published>2024-10-03T01:00:00Z</published><updated>2024-10-03T01:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/notes/rs-ecorisk-day1</id><summary type="html"><p>I'm at the Royal Society this morning for the 2 day programme on <a href="https://royalsociety.org/science-events-and-lectures/2024/10/ecological-and-commercial-risk/">&quot;How does ecological risk related to commercial risk?&quot;</a>, and am reporting on the <a href="https://royalsociety.org/-/media/events/2024/10/ecological-risk/programme-booklet.pdf">morning session</a>. The full program is being <a href="https://www.youtube.com/watch?v=gVuxzand8RE">livestreamed</a> so please do dial in if the below notes seem interesting to you. I put this note up almost live, so any errors below are my own.<em>(Update: partial <a href="https://anil.recoil.org/#daytwo">day 2 notes</a> now available below)</em></p><p>The summit kicked off with a keynote by economist <a href="https://en.wikipedia.org/wiki/Partha_Dasgupta">Sir Partha Dasgupta</a>. The focus was on the intersection of nature and economics, covering how markets fail to account for the ecosystems that sustain them. His <a href="https://www.gov.uk/government/publications/final-report-the-economics-of-biodiversity-the-dasgupta-review">landmark report</a> covered ecosystem services, freshwater, tipping points, and physical risk, bringing to light the urgent need to reframe economic activities around the services provided by nature.</p>···<p>As you can see, it's a very mixed bag of solutions I've come up with as I try to keep ownership of my Internet presence. I make not attempt to suggest any of this is optimal - more it's just a mix of pragmatism based on what I need versus how much effort I'm willing to put into it. But still, I see people talking about self hosting often, and so perhaps this'll both show you some options, and normalise the idea that it doesn't need to be perfect or all-or-nothing for you to make some inroads to taking back control of how you exist on the Internet. I don't use Azure or AWS for instance either, for a while I did, but I don't really operate at the scale where the complexity is justified - I'm not really dealing with a lot of traffic for most things. I don't deal with CDNs like Cloudflare either, as I'm not that big, and the few people who do want to see my websites probably don't care if it's a few seconds slower than it could be.</p><p>I'll perhaps try to do a follow on post in a couple of years to see how much, if anything has changed.</p></summary><category term="Digital Flapjack"></category></entry><entry><title>New set of ideas for incoming students</title><link href="https://anil.recoil.org/notes/student-ideas" rel="alternate"></link><published>2024-07-15T01:00:00Z</published><updated>2024-07-15T01:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/notes/student-ideas</id><summary type="html"><p>I've refreshed the set of project <a href="https://anil.recoil.org/ideas">ideas</a> for incoming <a href="https://www.cst.cam.ac.uk/teaching/part-ii">CST Part II</a> and <a href="https://www.cst.cam.ac.uk/teaching/masters">MPhil</a> and PhD student projects for 2024-2025.</p>-<p>These are not an exhaustive list, but intended to kickstart conversations for things we could work on together. Do get in touch if you're an incoming student and see something that grabs your interest.</p></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Mon 8 Jul 2024</title><link href="https://ryan.freumh.org/2024-07-08.html" rel="alternate"></link><published>2024-07-08T01:00:00Z</published><updated>2024-07-08T01:00:00Z</updated><author><name>RyanG</name></author><id>https://ryan.freumh.org/2024-07-08.html</id><summary type="html"><article>+<p>These are not an exhaustive list, but intended to kickstart conversations for things we could work on together. Do get in touch if you're an incoming student and see something that grabs your interest.</p></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Real-time Wildlife Monitoring</title><link href="https://watch.eeg.cl.cam.ac.uk/w/tyPqbNvp3isgTDZVVoLFD1" rel="alternate"></link><published>2024-07-12T19:52:48Z</published><updated>2024-07-12T19:52:48Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/tyPqbNvp3isgTDZVVoLFD1</id><summary type="html"><p>Abstract:<br />+This research introduces an AI-based alert system to reduce human-wildlife conflicts in the Romanian Carpathian Mountains. Globally, conflicts between people and wildlife are rising due to population growth, shifting land use patterns and climate change. In Romania, mountain communities are impacted by bears and wild boars, which damage livestock, crops and property. These conflicts can undermine conservation efforts and may result in the killing of problematic animals. In collaboration with Fundația Conservation Carpathia, this research supports Rapid Intervention Teams who respond to wildlife activity in mountain villages. Six years of camera trap data are used to train and test AI models to detect and classify European mammals. These models are integrated into an alert system and deployed in three locations. The new pipeline improves on the state-of-the-art for detecting and classifying bears and wild boars. Preliminary results from the field deployment show a positive impact on conservation efforts. This is the first known study to use remote processing of 4G-enabled camera trap images to operate a human-wildlife conflict alert system, with potential wider applications as cellular connectivity expands to more remote locations.</p>+Tom is an MRes student on the AI for Environmental Risk Centre for Doctoral Training at the University for Cambridge. He previously spent 10 years working for the UK's Foreign, Commonwealth and Development Office, where he designed and managed sustainable development projects while on postings in DRC, Sierra Leone and Tanzania.</p></summary><category term="EEG Seminars"></category></entry><entry><title>Mon 8 Jul 2024</title><link href="https://ryan.freumh.org/2024-07-08.html" rel="alternate"></link><published>2024-07-08T01:00:00Z</published><updated>2024-07-08T01:00:00Z</updated><author><name>RyanG</name></author><id>https://ryan.freumh.org/2024-07-08.html</id><summary type="html"><article><span> Previous: <a href="2024-06-24.html">Mon 24 Jun 2024</a> </span><span> Next: <a href="2024-08-05.html">Mon 5 Aug 2024</a> </span>···<img alt="" class="content-image" src="https://anil.recoil.org/images/compass24/compass24-22.webp" title="Made it back to London in time to catch some tennis" />-</p></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Second preprint of the LIFE biodiversity metric available</title><link href="https://anil.recoil.org/news/2024-life-2" rel="alternate"></link><published>2024-07-01T01:00:00Z</published><updated>2024-07-01T01:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/news/2024-life-2</id><summary type="html"><p>We have made an update to the <a href="https://anil.recoil.org/projects/life">LIFE</a> biodiversity metric based on reviewer feedback, and are very please that it has been accepted for publication early next year as part of a special issue from the Royal Society. Any comments would be most welcome before we submit the final proofs in a few months.</p>+</p></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Grid-friendly Energy Community Coordination for Reducing Grid Reinforcement Needs</title><link href="https://watch.eeg.cl.cam.ac.uk/w/iSPamqxUdmP2CwNNdGyQSN" rel="alternate"></link><published>2024-07-05T18:14:38Z</published><updated>2024-07-05T18:14:38Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/iSPamqxUdmP2CwNNdGyQSN</id><summary type="html"><p>Abstract:<br />+This talk discusses how market mechanisms and automated trading strategies can be used to control the flexible consumption and generation units of the community members in such a way that they make the best possible use of existing distribution networks and support the network operator in avoiding and eliminating congestion situations. This ultimately helps avoiding grid reinforcements or allows to provide a better service with the existing grid, keeping in mind that it takes much longer to reinforce the grid than to build and connect many new (fluctuating) decentralized renewable generators and new loads such as heat pumps and electric vehicles.</p>+Since 2017: Professor of Control and Integration of Grids at INATECH; before: Professor for Energy Systems Technology and Energy Economics, in particular intelligent decentralized structures for sustainable power supply (Smart Grids) at Offenburg University of Applied Sciences, Fellow and head of the research project “Smart Grids” at the foundation neue verantwortung, Berlin, Senior Researcher and Project Manager in the research area “Future Energy Systems”, SAP AG, Research Assistant at the University of Mannheim, Research Fellow at the Iowa State University, Research Assistant at the University of Karlsruhe (TH)</p></summary><category term="EEG Seminars"></category></entry><entry><title>Second preprint of the LIFE biodiversity metric available</title><link href="https://anil.recoil.org/news/2024-life-2" rel="alternate"></link><published>2024-07-01T01:00:00Z</published><updated>2024-07-01T01:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/news/2024-life-2</id><summary type="html"><p>We have made an update to the <a href="https://anil.recoil.org/projects/life">LIFE</a> biodiversity metric based on reviewer feedback, and are very please that it has been accepted for publication early next year as part of a special issue from the Royal Society. Any comments would be most welcome before we submit the final proofs in a few months.</p>···&nbsp; <a href="https://anil.recoil.org/papers/2024-life.pdf"><span class="nobreak">PDF<img alt="pdf" class="inline-icon" src="https://anil.recoil.org/assets/pdf.svg" /></span></a>-</blockquote></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Reinstalling `maestral`</title><link href="https://mort.io/blog/reinstall-maestral/" rel="alternate"></link><published>2024-06-28T01:00:00Z</published><updated>2024-06-28T01:00:00Z</updated><author><name>RichardM</name></author><id>https://mort.io/blog/reinstall-maestral/</id><summary type="html"><p>A short one this; just a crib of commands to redo <code>maestral</code> (unofficial+</blockquote></summary><category term="Anil Madhavapeddy's feed"></category></entry><entry><title>Annual Monitoring of Forest AGB</title><link href="https://watch.eeg.cl.cam.ac.uk/w/2kgatqvAtb6XUAgF2UoKZz" rel="alternate"></link><published>2024-06-28T18:13:31Z</published><updated>2024-06-28T18:13:31Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/2kgatqvAtb6XUAgF2UoKZz</id><summary type="html"><p>Title:<br />+Annual Monitoring of Forest AGB over a Period of 10 years Using SSL-derived Representations from Optical Time Series</p>+I recap the functioning of our fully self-supervised learning pipeline based on the spectral-temporal Barlow Twins. The SSL approach generates highly informative representations at 10m spatial resolution from cloud-corrupted optical time series. The resulting representations are well correlated with GEDI-derived relative height measurements so that an AGB model for vegetation/forest of up to 300-500 t/ha can be derived. I show that the model transfers well between years making it possible to train the model on (for example) one year of Sentinel-2 data together with the corresponding GEDI measurements, and applying the frozen model to Landsat data acquired in previous years.</p>+2010-2023: Full Professor for Remote Sensing/Geomatics - since 2016 Lead of Mantle's research team.</p></summary><category term="EEG Seminars"></category></entry><entry><title>HyWay: Enabling Mingling in the Hybrid World</title><link href="https://watch.eeg.cl.cam.ac.uk/w/pUULKS4bi9hG9uUy9prhF2" rel="alternate"></link><published>2024-06-28T18:13:11Z</published><updated>2024-06-28T18:13:11Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/pUULKS4bi9hG9uUy9prhF2</id><summary type="html"><p>We present HyWay, short for “Hybrid Hallway”, to enable mingling and informal interaction among in-person and remote users in semi-structured and unstructured hybrid settings. Key to the design of HyWay is bridging the awareness gap between physical and virtual users and providing the virtual users the same agency as physical users. We explain how HyWay achieves these goals with an agile, light touch approach that centers on commodity hardware, frictionless bootstrapping, and a deploy-learn-refine flywheel to improve the system continually. We present our learnings from multiple deployments and discuss ongoing research aimed at weaving together more tightly the virtual and the physical realms.</p>+Venkat Padmanabhan is the Managing Director of Microsoft Research India in Bengaluru. He was previously with Microsoft Research Redmond, USA for nearly 9 years. Venkat’s research interests are broadly in networked and mobile computing systems, and his work over the years has led to highly cited papers and paper awards, technology transfers within Microsoft, and industry impact. He has received several awards and recognitions, including the Shanti Swarup Bhatnagar Prize in 2016, four test-of-time paper awards from ACM SIGMOBILE, ACM SIGMM, and ACM SenSys, and several best paper awards. He was also among those recognized with the ACM SIGCOMM Networking Systems Award 2020, for contributions to the ns family of network simulators. Venkat holds a B.Tech. from IIT Delhi (from where he received the Distinguished Alumnus award in 2018) and an M.S. and a Ph.D. from UC Berkeley, all in Computer Science, and has been elected a Fellow of the INAE, the IEEE, and the ACM. He is an adjunct professor at the Indian Institute of Science and was previously an affiliate faculty member at the University of Washington.</p></summary><category term="EEG Seminars"></category></entry><entry><title>Reinstalling `maestral`</title><link href="https://mort.io/blog/reinstall-maestral/" rel="alternate"></link><published>2024-06-28T01:00:00Z</published><updated>2024-06-28T01:00:00Z</updated><author><name>RichardM</name></author><id>https://mort.io/blog/reinstall-maestral/</id><summary type="html"><p>A short one this; just a crib of commands to redo <code>maestral</code> (unofficial<pre class="language-bash z-code"><code class="language-bash"><span class="z-source z-shell z-bash"><span class="z-comment z-line z-number-sign z-shell"><span class="z-punctuation z-definition z-comment z-begin z-shell">#</span></span><span class="z-comment z-line z-number-sign z-shell"> remove existing configuration</span><span class="z-comment z-line z-number-sign z-shell"></span></span><span class="z-source z-shell z-bash"><span class="z-meta z-function-call z-shell"><span class="z-variable z-function z-shell">rm</span></span><span class="z-meta z-function-call z-arguments z-shell"> <span class="z-meta z-group z-expansion z-tilde"><span class="z-variable z-language z-tilde z-shell">~</span></span>/.config/maestral/maestral.ini</span>···-</article></summary><category term="ryan.freumh.org"></category></entry><entry><title>Interview with AI@CAM about conservation</title><link href="https://anil.recoil.org/notes/aicam-interview-ce" rel="alternate"></link><published>2024-06-09T01:00:00Z</published><updated>2024-06-09T01:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/notes/aicam-interview-ce</id><summary type="html"><p>I talked to the <a href="https://ai.cam.ac.uk">AI@Cam</a> team to discuss our <a href="https://anil.recoil.org/notes/aicn-in-aicam">AICN</a>+</article></summary><category term="ryan.freumh.org"></category></entry><entry><title>Tackling the Hidden Costs of Computational Science</title><link href="https://watch.eeg.cl.cam.ac.uk/w/nfXCsAiyawBPM2qzdpRCfB" rel="alternate"></link><published>2024-06-21T16:03:59Z</published><updated>2024-06-21T16:03:59Z</updated><author><name>Seminars</name></author><id>https://watch.eeg.cl.cam.ac.uk/w/nfXCsAiyawBPM2qzdpRCfB</id><summary type="html"><p>Title:<br />+Tackling the Hidden Costs of Computational Science: GREENER Principles for Environmentally Sustainable Research</p>+From genetic studies and astrophysics simulations to AI, scientific computing has enabled amazing discoveries and there is no doubt it will continue to do so. However, the corresponding energy usage and environmental impacts are a growing concern in light of the urgency of the climate crisis, so what can we all do about it? Tackling this issue and making it easier for scientists to engage with sustainable computing is what motivated the Green Algorithms project. Through the prism of the GREENER principles for environmentally sustainable science, we will discuss what we learned along the way, how to estimate the impact of our work and what levers scientists and institutions have to make their research more sustainable. We will also debate what hurdles exist and what is still needed moving forward.</p>+Dr Loïc Lannelongue is a Research Associate in Biomedical Data Science in the Heart and Lung Research Institute at the University of Cambridge, UK, and the Cambridge-Baker Systems Genomics Initiative. He leads the Green Algorithms project, an initiative promoting more environmentally sustainable computational science. His research interests also include radiogenomics, i.e. combining medical imaging and genetic information with machine learning to better understand and treat cardiovascular diseases. He is a Software Sustainability Institute Fellow, a Post-doctoral Associate at Jesus College, Cambridge, and an Associate Fellow of the Higher Education Academy.</p></summary><category term="EEG Seminars"></category></entry><entry><title>Interview with AI@CAM about conservation</title><link href="https://anil.recoil.org/notes/aicam-interview-ce" rel="alternate"></link><published>2024-06-09T01:00:00Z</published><updated>2024-06-09T01:00:00Z</updated><author><name>AnilM</name></author><id>https://anil.recoil.org/notes/aicam-interview-ce</id><summary type="html"><p>I talked to the <a href="https://ai.cam.ac.uk">AI@Cam</a> team to discuss our <a href="https://anil.recoil.org/notes/aicn-in-aicam">AICN</a>project and what we're planning to do in the <a href="https://anil.recoil.org/projects/ce">Conservation Evidence Copilots</a> team.</p><p>Over the last two decades, the University of Cambridge-based project Conservation Evidence has screened more than 1.6 million scientific papers on conservation, as well as manually summarising 8,600+ studies relating to conservation actions. However, the current project’s work is limited by the specialised skills needed to screen and summarise relevant studies. It took more than 75 person years to manually curate the current database and only a few 100 papers can be added each year. By accelerating these efforts, AI has the potential to transform the impact this database has on biodiversity conservation.</p>
+3
feed.json
+3
feed.json
+19
-3
index.html
+19
-3
index.html
············else if (url.hostname === 'bsky.app' || url.hostname === 'bsky.social') linkType = 'bluesky';return `<a href="${link.url}" target="_blank" class="external-link-item" title="${link.url}" data-link-type="${linkType}">${displayText}</a>`;······
+30
-14
make_threads.py
+30
-14
make_threads.py
············original_link = entry['content_links'][i] if i < len(entry['content_links']) else normalized_link······
+4
mapping.json
+4
mapping.json
+209
-4248
threads.json
+209
-4248
threads.json
······"url": "https://git.sr.ht/~jonsterling/forester-base-theme/commit/a251f9cf19b0ff42f4553d315df5181b985c79cb","normalized_url": "https://git.sr.ht/~jonsterling/forester-base-theme/commit/a251f9cf19b0ff42f4553d315df5181b985c79cb",·········"url": "https://www.abebooks.co.uk/9789812701428/Domain-theoretic-Foundations-Functional-Programming-Streicher-9812701427/plp","normalized_url": "https://abebooks.co.uk/9789812701428/Domain-theoretic-Foundations-Functional-Programming-Streicher-9812701427/plp",····································"url": "https://defraenvironment.blog.gov.uk/2025/01/20/newly-recognised-official-statistic-tracks-the-environmental-impact-of-our-consumption/","normalized_url": "https://defraenvironment.blog.gov.uk/2025/01/20/newly-recognised-official-statistic-tracks-the-environmental-impact-of-our-consumption",······"url": "https://github.com/RyanGibb/babel/commit/c975ab938efb1330a691e4883bc1e732b21cddf3","normalized_url": "https://github.com/RyanGibb/babel/commit/c975ab938efb1330a691e4883bc1e732b21cddf3",···"title": "A fully AI-generated paper just passed peer review; notes from our evidence synthesis workshop",·····················"url": "https://github.com/ocaml-multicore/effects-examples/blob/master/multishot/memo.ml","normalized_url": "https://github.com/ocaml-multicore/effects-examples/blob/master/multishot/memo.ml",······"url": "https://github.com/ai-robots-txt/ai.robots.txt/blob/main/table-of-bot-metrics.md","normalized_url": "https://github.com/ai-robots-txt/ai.robots.txt/blob/main/table-of-bot-metrics.md",···············"normalized_url": "https://royalsociety.org/news-resources/projects/science-in-the-age-of-ai",············"url": "https://blog.hillion.co.uk/posts/void-processes/dissertation/jsh77-dissertation.pdf","normalized_url": "https://blog.hillion.co.uk/posts/void-processes/dissertation/jsh77-dissertation.pdf",··················"url": "https://www.docker.com/blog/docker-index-dramatic-growth-in-docker-usage-affirms-the-continued-rising-power-of-developers/","normalized_url": "https://docker.com/blog/docker-index-dramatic-growth-in-docker-usage-affirms-the-continued-rising-power-of-developers",·································"url": "https://www.bbc.com/future/article/20241011-what-explains-increasing-anxiety-about-ultra-processed-plant-based-foods","normalized_url": "https://bbc.com/future/article/20241011-what-explains-increasing-anxiety-about-ultra-processed-plant-based-foods",·········"url": "https://www.kings.cam.ac.uk/news/2023/kings-unveils-new-solar-panels-restored-chapel-roof","normalized_url": "https://kings.cam.ac.uk/news/2023/kings-unveils-new-solar-panels-restored-chapel-roof",······"url": "https://www.cshss.cam.ac.uk/research-info/j-m-keynes-fellowship-fund/j-m-keynes-fellows","normalized_url": "https://cshss.cam.ac.uk/research-info/j-m-keynes-fellowship-fund/j-m-keynes-fellows",······························-"url": "https://ryan.freumh.org/~/.config/emacs/modules/lang/org/config.el::(map! :map evil-org-mode-map",-"normalized_url": "https://ryan.freumh.org/~/.config/emacs/modules/lang/org/config.el::(map! :map evil-org-mode-map",···"url": "https://www.gov.uk/government/publications/revenue-and-customs-brief-7-2024-vat-treatment-of-voluntary-carbon-credits/revenue-and-customs-brief-vat-treatment-of-voluntary-carbon-credits","normalized_url": "https://gov.uk/government/publications/revenue-and-customs-brief-7-2024-vat-treatment-of-voluntary-carbon-credits/revenue-and-customs-brief-vat-treatment-of-voluntary-carbon-credits",··················"url": "https://www.vice.com/en/article/the-amazon-is-worth-more-money-left-standing-study-shows/","normalized_url": "https://vice.com/en/article/the-amazon-is-worth-more-money-left-standing-study-shows","url": "https://shopping.rspb.org.uk/gifts-home/home-and-kitchen/food-drink/food/gola-chocolate.html","normalized_url": "https://shopping.rspb.org.uk/gifts-home/home-and-kitchen/food-drink/food/gola-chocolate.html",······"url": "https://www.xilinx.com/publications/product-briefs/xilinx-smartssd-computational-storage-drive-product-brief.pdf","normalized_url": "https://xilinx.com/publications/product-briefs/xilinx-smartssd-computational-storage-drive-product-brief.pdf",···············"normalized_url": "https://cambridge.org/engage/coe/article-details/679385946dde43c9082f7009",····································-"normalized_url": "https://digitalflapjack.com/blog/the-partially-dynamic-web/https/ocaml.org",·········"normalized_url": "https://sicsa.ac.uk/wp-content/uploads/2024/11/LOCO2024_paper_35.pdf",···"normalized_url": "https://sicsa.ac.uk/wp-content/uploads/2024/11/LOCO2024_paper_30.pdf",···"normalized_url": "https://sicsa.ac.uk/wp-content/uploads/2024/11/LOCO2024_paper_28.pdf",···"normalized_url": "https://sicsa.ac.uk/wp-content/uploads/2024/11/LOCO2024_paper_30.pdf",································································································"url": "https://www.linkedin.com/posts/sophus-zu-ermgassen-12915ba6_nature-based-carbon-markets-have-experienced-activity-7226538933961007104-mM-u?utm_source=share&utm_medium=member_desktop","normalized_url": "https://linkedin.com/posts/sophus-zu-ermgassen-12915ba6_nature-based-carbon-markets-have-experienced-activity-7226538933961007104-mM-u",···"url": "https://ark.intel.com/content/www/us/en/ark/products/231803/intel-processor-n100-6m-cache-up-to-3-40-ghz.html","normalized_url": "https://ark.intel.com/content/www/us/en/ark/products/231803/intel-processor-n100-6m-cache-up-to-3-40-ghz.html",······························"url": "https://docs.google.com/presentation/d/1ZZMqF-8hCIupNm5VUH8wu61v9eTuI1e-/edit#slide=id.p1","normalized_url": "https://docs.google.com/presentation/d/1ZZMqF-8hCIupNm5VUH8wu61v9eTuI1e-/edit",···+"title": "Grid-friendly Energy Community Coordination for Reducing Grid Reinforcement Needs",·········"url": "https://about.conservationevidence.com/2024/07/12/the-next-steps-for-transforming-conservation-ideas-from-the-effectiveness-revolution-workshop/","normalized_url": "https://about.conservationevidence.com/2024/07/12/the-next-steps-for-transforming-conservation-ideas-from-the-effectiveness-revolution-workshop",···························"normalized_url": "https://cl.cam.ac.uk/research/dtg/attarchive/pub/docs/att/tr.94.4.pdf",···"normalized_url": "https://gitlab.freedesktop.org/wlroots/wlroots/-/blob/master/tinywl/tinywl.c",·································"url": "https://fosdem.org/2024/schedule/event/fosdem-2024-3242-screen-sharing-on-raspberry-pi-5-using-vnc-in-weston-and-wayland-with-the-yocto-project-and-openembedded/","normalized_url": "https://fosdem.org/2024/schedule/event/fosdem-2024-3242-screen-sharing-on-raspberry-pi-5-using-vnc-in-weston-and-wayland-with-the-yocto-project-and-openembedded",·········"normalized_url": "https://undonecs.sciencesconf.org/data/Undonecs_2024_abstract_43.pdf",······-"url": "https://digitalflapjack.com/blog/a-journey-into-minimal-computing-or-my-slow-evolving-2021-side-project/",-"normalized_url": "https://digitalflapjack.com/blog/a-journey-into-minimal-computing-or-my-slow-evolving-2021-side-project",························"url": "https://www.conservation.cam.ac.uk/events/online-workshop-delivering-effectiveness-revolution-conservation-lessons-organisations-policy","normalized_url": "https://conservation.cam.ac.uk/events/online-workshop-delivering-effectiveness-revolution-conservation-lessons-organisations-policy",············"url": "https://www.usenix.org/system/files/conference/usenixsecurity17/sec17-chung.pdf","normalized_url": "https://usenix.org/system/files/conference/usenixsecurity17/sec17-chung.pdf",············-"normalized_url": "https://ryan.freumh.org/matrix-synapse-authentication-with-raven-shibboleth","url": "https://matrix-org.github.io/synapse/latest/usage/configuration/config_documentation.html#saml2_config","normalized_url": "https://matrix-org.github.io/synapse/latest/usage/configuration/config_documentation.html",·································"url": "https://icfp23.sigplan.org/details/ocaml-2023-papers/15/State-of-the-OCaml-Platform-2023","normalized_url": "https://icfp23.sigplan.org/details/ocaml-2023-papers/15/State-of-the-OCaml-Platform-2023",···"url": "https://icfp23.sigplan.org/details/ocaml-2023-papers/5/Eio-1-0-Effects-based-IO-for-OCaml-5","normalized_url": "https://icfp23.sigplan.org/details/ocaml-2023-papers/5/Eio-1-0-Effects-based-IO-for-OCaml-5",······-"url": "https://digitalflapjack.com/blog/some-notes-on-processing-and-display-geospatial-data/",-"normalized_url": "https://digitalflapjack.com/blog/some-notes-on-processing-and-display-geospatial-data",··································································"normalized_url": "https://linkedin.com/feed/update/urn:li:activity:7228682518596603904",············"url": "https://news.mongabay.com/2025/01/life-scores-map-out-where-habitat-loss-for-crops-drives-extinction/","normalized_url": "https://news.mongabay.com/2025/01/life-scores-map-out-where-habitat-loss-for-crops-drives-extinction",···············"normalized_url": "https://tarides.com/blog/2022-12-14-hillingar-mirageos-unikernels-on-nixos",············"url": "https://icfp22.sigplan.org/details/ocaml-2022-papers/8/Homogeneous-builds-with-OBuilder-and-OCaml","normalized_url": "https://icfp22.sigplan.org/details/ocaml-2022-papers/8/Homogeneous-builds-with-OBuilder-and-OCaml",······························"url": "https://www.pem.cam.ac.uk/college/corporate-partnership/25th-anniversary-corporate-partnership-programme/25th-anniversary-11","normalized_url": "https://pem.cam.ac.uk/college/corporate-partnership/25th-anniversary-corporate-partnership-programme/25th-anniversary-11",·····················"url": "https://icfp21.sigplan.org/details/ocaml-2021-papers/16/Experiences-with-Effects","normalized_url": "https://icfp21.sigplan.org/details/ocaml-2021-papers/16/Experiences-with-Effects",············"url": "https://studres.cs.st-andrews.ac.uk/Library/ProjectLibrary/cs4099/2021/rtg2-Final_Report.pdf","normalized_url": "https://studres.cs.st-andrews.ac.uk/Library/ProjectLibrary/cs4099/2021/rtg2-Final_Report.pdf",···························"url": "https://github.com/ocaml-multicore/multicore-talks/tree/master/ocaml2020-workshop-parallel","normalized_url": "https://github.com/ocaml-multicore/multicore-talks/tree/master/ocaml2020-workshop-parallel",···"url": "https://icfp20.sigplan.org/details/ocaml-2020-papers/6/OCaml-CI-A-Zero-Configuration-CI","normalized_url": "https://icfp20.sigplan.org/details/ocaml-2020-papers/6/OCaml-CI-A-Zero-Configuration-CI",······"normalized_url": "https://discuss.ocaml.org/t/multicore-ocaml-january-2020-update/5090",·····································································"url": "https://github.com/mor1/ocal/blob/13a9a7f5b8f2e0be4c2b55941a00a885df202cf8/ocal.opam#L16-L22","normalized_url": "https://github.com/mor1/ocal/blob/13a9a7f5b8f2e0be4c2b55941a00a885df202cf8/ocal.opam",·········"normalized_url": "https://usenix.org/conference/atc16/technical-sessions/presentation/alim",···························"url": "https://www.usenix.org/conference/usenixsecurity15/technical-sessions/presentation/kaloper-mersinjak","normalized_url": "https://usenix.org/conference/usenixsecurity15/technical-sessions/presentation/kaloper-mersinjak",············"url": "https://www.usenix.org/conference/nsdi15/technical-sessions/presentation/madhavapeddy","normalized_url": "https://usenix.org/conference/nsdi15/technical-sessions/presentation/madhavapeddy",·········-"normalized_url": "https://anil.recoil.org/2014/04/08/grepping-every-known-ocaml-package-source.html",·····················"normalized_url": "http://lists.ocaml.org/pipermail/opam-devel/2015-February/000940.html",··················"url": "https://www.cambridge.org/core/journals/journal-of-functional-programming/article/cufp13-scribes-report/F38AAE60DA9AD95E1737E3F863075C13","normalized_url": "https://cambridge.org/core/journals/journal-of-functional-programming/article/cufp13-scribes-report/F38AAE60DA9AD95E1737E3F863075C13",······················································-"normalized_url": "https://anil.recoil.org/2013/09/16/camlpdf-the-end-of-sucky-pdf-tools.html",·············································"url": "https://www.cambridge.org/core/journals/journal-of-functional-programming/article/commercial-users-of-functional-programming-workshop-report/7B8E5D99E6C0D40D45B37D972B82598D","normalized_url": "https://cambridge.org/core/journals/journal-of-functional-programming/article/commercial-users-of-functional-programming-workshop-report/7B8E5D99E6C0D40D45B37D972B82598D",··················"url": "https://github.com/ocaml/v2.ocaml.org/blob/master/site/meetings/ocaml/2013/proposals/ocamlot.pdf","normalized_url": "https://github.com/ocaml/v2.ocaml.org/blob/master/site/meetings/ocaml/2013/proposals/ocamlot.pdf",···"normalized_url": "https://usenix.org/conference/foci13/workshop-program/presentation/rotsos",····························································"url": "https://discuss.ocaml.org/t/v3-ocaml-org-a-roadmap-for-ocamls-online-presence/8368/18","normalized_url": "https://discuss.ocaml.org/t/v3-ocaml-org-a-roadmap-for-ocamls-online-presence/8368/18",···"url": "https://www.youtube.com/watch?v=dEUMNuE4rxc&list=PLnqUlCo055hU46uoONmhYGUbYAK27Y6rS&index=8","normalized_url": "https://youtube.com/watch?v=dEUMNuE4rxc&list=PLnqUlCo055hU46uoONmhYGUbYAK27Y6rS&index=8",····················································································"url": "https://www.usenix.org/conference/hotcloud-10/turning-down-lamp-software-specialisation-cloud","normalized_url": "https://usenix.org/conference/hotcloud-10/turning-down-lamp-software-specialisation-cloud",····························································"normalized_url": "https://web.archive.org/web/20081121042533/https://xenserver5.com/ha.php",································································································"url": "https://www.nytimes.com/2004/10/07/technology/circuits/connecting-paper-and-online-worlds-by-cellphone-camera.html","normalized_url": "https://nytimes.com/2004/10/07/technology/circuits/connecting-paper-and-online-worlds-by-cellphone-camera.html",·································