Reposted by Levi LelisGaia Vince @wanderinggaia.bsky.social · 11/09/2025Brazil shows how it’s done in a democracytheguardian.comBrazil’s supreme court finds Bolsonaro guilty of plotting military coupFormer president faces decades-long jail sentence for seeking to forcibly cling to power after losing 2022 election 530982
Reposted by Levi LelisMatthew Guzdial @matthewguz.bsky.social · 25/08/2025Excited to announce that our work on Reinforcement Learning for Arachnophobia treatment has been accepted at ACM Transactions on Interactive Intelligent Systems! We found that an RL agent could more effectively adapt VR spiders to achieve specified anxiety levels in users compared to current SOTA. 5567
Reposted by Levi LelisEugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 27/07/2025Was talking to a student who wasn't sure about why one would get a PhD. So I wrote up a list of reasons! www.eugenevinitsky.com/posts/reason...eugenevinitsky.comEugene Vinitsky 75111
Reposted by Levi LelisLevi Lelis @programsynthesis.bsky.social · 02/07/2025Previous work has shown that programmatic policies—computer programs written in a domain-specific language—generalize to out-of-distribution problems more easily than neural policies. Is this really the case? 🧵 294
Levi Lelis @programsynthesis.bsky.social · 02/07/2025Previous work has shown that programmatic policies—computer programs written in a domain-specific language—generalize to out-of-distribution problems more easily than neural policies. Is this really the case? 🧵 294
Reposted by Levi LelisMarc Lanctot @sharky6000.bsky.social · 29/06/2025If like me your Discover feed has been even worse lately and you are here for ML/AI news and discussion, check out these two feeds: - Paper Skygest - ML Feed: Trending Links below 👇 3324
Reposted by Levi LelisMartin Klissarov @martinklissarov.bsky.social · 27/06/2025As AI agents face increasingly long and complex tasks, decomposing them into subtasks becomes increasingly appealing. But how do we discover such temporal structure? Hierarchical RL provides a natural formalism-yet many questions remain open. Here's our overview of the field🧵 13510
Reposted by Levi LelisMark Gongloff @markgongloff.bsky.social · 24/06/2025As hot as this summer is, it’s also one of the coolest we’ll ever enjoy again. Just how much hotter and deadlier summers will get is still up to us. Right now we’re working hard to make them worse 🎁 link to my @opinion.bloomberg.com column: www.bloomberg.com/opinion/arti...bloomberg.comThe Heat Dome Wants a Word With Climate-Change DeniersThe temperatures gripping the US this week were made up to five times more likely by the fact that the atmosphere is simply hotter. 38440
Reposted by Levi LelisEugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 21/06/2025Hiring a postdoc to scale up and deploy RL-based planning onto some self-driving cars! We'll be building on arxiv.org/abs/2502.03349 and learn what the limits and challenges of RL planning are. Shoot me a message if interested and help spread the word please! Full posting to come in a bit.arxiv.orgRobust Autonomy Emerges from Self-PlaySelf-play has powered breakthroughs in two-player and multi-player games. Here we show that self-play is a surprisingly effective strategy in another domain. We show that robust and naturalistic drivi... 36025
Reposted by Levi LelisMatthew Guzdial @matthewguz.bsky.social · 16/06/2025We’re extending the AIIDE deadline! Partially due to author requests, partially due to a significant increase in submissions meaning I need to increase the PC! 154
Levi Lelis @programsynthesis.bsky.social · 13/06/2025🧵1/ New paper! 📄 Subgoal-Guided Policy Heuristic Search with Learned Subgoals, led by my PhD student @tuero.ca. arxiv.org/pdf/2506.07255 This paper follows the Levin tree search (LTS) research line and focuses on learning subgoal-based policies. 153
Levi Lelis @programsynthesis.bsky.social · 12/06/2025I enjoyed reading this and the comments within it. 000
Reposted by Levi LelisMatthew Guzdial @matthewguz.bsky.social · 06/06/2025After literal years of being down due to security issues, my blog is now back up, including my one post on "How to Write an AIIDE Paper" www.guzdial.com/blog/how-to-...guzdial.comHow to Write an AIIDE Paper – Matthew Guzdial Blog 072
Reposted by Levi LelisJulian Togelius @togelius.bsky.social · 03/06/2025We are very happy to report that the second edition of our textbook on Artificial Intelligence and Games is now finally published! This book is a thorough update of our popular textbook, trying to provide a comprehensive coverage of the many aspects of and use cases for AI in games. 3325
Reposted by Levi LelisMatthew Guzdial @matthewguz.bsky.social · 30/05/2025Excited to announce that the second edition of the PCGML textbook by @sampsnodgrass.bsky.social, @autumnsburg.bsky.social, and myself is up! This version is a major update of the first, with a new chapter on GenAI and updates on the last two years of research. link.springer.com/book/10.1007...link.springer.comProcedural Content Generation via Machine LearningThis book updates and expands upon the first beginner-focused guide to procedural content generation via machine learning (PCGML) 071
Levi Lelis @programsynthesis.bsky.social · 28/05/2025We have extended the submission deadlines for these workshops. Programmatic Representations for Agent Learning (ICML) - Vancouver. - Deadline: May 30, 2025 (pral-workshop.github.io) Programmatic Reinforcement Learning (RLC) - Edmonton. - Deadline: June 6, 2025 (prl-workshop.github.io)pral-workshop.github.ioProgrammatic Representations for Agent Learning Workshop at ICML 2025 041
Reposted by Levi LelisMarlos C. Machado @marloscmachado.bsky.social · 27/05/2025📢 I'm very excited to release AgarCL, a new evaluation platform for research in continual reinforcement learning‼️ Repo: github.com/machado-rese... Website: agarcl.github.io Preprint: arxiv.org/abs/2505.18347 Details below 👇 1298
Reposted by Levi LelisLevi Lelis @programsynthesis.bsky.social · 23/05/2025🧵1/ New paper! 📄 InnateCoder: Learning Programmatic Options with Foundation Models This is Rubens Moraes' final chapter of his PhD thesis from Universidade Federal de Viçosa, Brazil, in collaboration with Quazi Sadmine and Hendrik Baier. arXiv: arxiv.org/abs/2505.12508 163
Reposted by Levi LelisMarlos C. Machado @marloscmachado.bsky.social · 24/05/2025📢 I'm happy to share the preprint: _Reward-Aware Proto-Representations in Reinforcement Learning_ ‼️ My PhD student, Hon Tik Tse, led this work, and my MSc student, Siddarth Chandrasekar, assisted us. arxiv.org/abs/2505.16217 Basically, it's the SR with rewards. See below 👇 24010
Levi Lelis @programsynthesis.bsky.social · 23/05/2025🧵1/ New paper! 📄 InnateCoder: Learning Programmatic Options with Foundation Models This is Rubens Moraes' final chapter of his PhD thesis from Universidade Federal de Viçosa, Brazil, in collaboration with Quazi Sadmine and Hendrik Baier. arXiv: arxiv.org/abs/2505.12508 163
Reposted by Levi LelisBenjamin Heymann @benhey.bsky.social · 23/05/2025🎉 Super excited: today @sharky6000.bsky.social is presenting our new algorithm Progressive Hiding at #AAMAS2025! It's a learning method for games with imperfect information. 🔗 See his post: bsky.app/profile/shar... Wish I could be there! 😢 1/6 174
Levi Lelis @programsynthesis.bsky.social · 21/05/2025Are you interested in programmatic representations in machine learning? Please consider submitting your work to these workshops: one at ICML and another at RLC. 🧵 260
Levi Lelis @programsynthesis.bsky.social · 17/05/2025Wow! Great list of papers on the computational capability of neural networks. 010
Reposted by Levi LelisRoxana Rădulescu @rroxana.bsky.social · 13/05/2025🎓 PhD position available! Join our interdisciplinary research project on causal agent-based modelling! 🔍 Looking for curious minds with a MSc degree (or near to completing one) in CS/AI/related fields. 📍 Location: Utrecht University, NL 🗓️ Deadline: 16 June 2025 📩 Info: www.uu.nl/en/organisat...uu.nlPhD Position in Causal Agent-based Modelling of Complex Social SystemsJoin this exciting interdisciplinary research project at the Centre for Complex Systems Studies and study causal agent-based modelling! 075
Reposted by Levi LelisMarc Lanctot @sharky6000.bsky.social · 08/05/2025A bit of an old paper but I'm still excited about it, I've never posted about it, and we just released code for the main algorithm: Population RL (PopRL). 👇 Population-based Evaluation in Repeated Rock-Paper-Scissors as a Benchmark for Multiagent Reinforcement Learning, accepted (TMLR '23) 🧵 1/N 1121
Levi Lelis @programsynthesis.bsky.social · 29/04/2025So happy to see that ICLR will go to Brazil next year! 010
Reposted by Levi LelisNathan Lambert @natolambert.bsky.social · 18/04/2025rlhfbook also available on arxiv for SEO 😀 happy friday arxiv.org/abs/2504.12501arxiv.orgReinforcement Learning from Human FeedbackReinforcement learning from human feedback (RLHF) has become an important technical and storytelling tool to deploy the latest machine learning systems. In this book, we hope to give a gentle… 36913
Reposted by Levi LelisDennis Soemers @dennissoemers.bsky.social · 18/04/2025We have a new open position for a PhD student to work on Reinforcement Learning for Cyber Security! Check out vacancies.maastrichtuniversity.nl/job/Maastric... for details, and feel free to contact me for any info as well. Application deadline: May 11 Reposts much appreciated. 065
Reposted by Levi LelisMark J. Nelson @mm-jj-nn.bsky.social · 07/04/2025Three postdoc positions in HCI & Videogame Theory at the IT University of Copenhagen. Deadline May 15. "Candidates will investigate current theory use in HCI/games research and practice, develop approaches to make theory more usable, and evaluate the effectiveness of such interventions"candidate.hr-manager.net Postdoc positions in HCI and Videogames Theory at IT University of Copenhagen 084
Reposted by Levi LelisAIIDE @aiide.bsky.social · 03/04/2025👾Call for Papers for #AIIDE25 is live! 👾 Join us in Edmonton, Canada where we will be celebrating over 20 years of AIIDE with a special theme: Strong Foundations! We encourage the submission of work that builds on prior work from AIIDE proceedings. 🔍Info in replies!sites.google.comAIIDE 2025 - Call for PapersAIIDE 2025 welcomes submissions across the vast field of Artificial Intelligence and Interactive Digital Entertainment. We are particularly interested in novel contributions and applications, as well ... 11011
Reposted by Levi LelisAIIDE @aiide.bsky.social · 17/03/2025We are very pleased to announce that AIIDE'25 will be back at the University of Alberta in Edmonton, Alberta, Canada from November 10-14! Paper deadline June 21 AOE. Join us with your work at the intersection of AI and Digital Entertainment! sites.google.com/ualberta.ca/...sites.google.comAIIDE 2025The 21st AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE 2025) is the next in an annual series of conferences showcasing interdisciplinary research on modeling,... 1610
Reposted by Levi LelisNicolas Papernot @nicolaspapernot.bsky.social · 12/03/2025The Canadian AI Safety Institute (CAISI) Research Program at CIFAR is now accepting Expressions of Interest for Solution Networks in AI Safety under two themes: * Mitigating the Safety Risks of Synthetic Content * AI Safety in the Global South. cifar.ca/ai/ai-and-so... 073
Reposted by Levi LelisGautam Kamath @gautamkamath.com · 11/03/2025There is a postdoctoral position for research on adversarial robustness at UBC, supervised by Mathias Lecuyer (@mathias-lecuyer.bsky.social), Geoff Pleiss, and Nidhi Hegde. Please spread the word! 🇨🇦 docs.google.com/forms/d/e/1F...docs.google.comApplication for a postdoctoral position on adversarial robustnessLocation - Work primarily takes place at UBC, Vancouver, Computer Science department. Position - We are recruiting a postdoctoral researcher for a funded position, under the joint supervision of Math... 1185
Levi Lelis @programsynthesis.bsky.social · 06/03/2025This recognition is long overdue. I was so happy to wake up to this news today. awards.acm.org/about/2024-t...awards.acm.orgAndrew Barto and Richard Sutton are the recipients of the 2024 ACM A.M. Turing Award for developing the conceptual and algorithmic foundations of reinforcement learning.Andrew Barto and Richard Sutton as the recipients of the 2024 ACM A.M. Turing Award for developing the conceptual and algorithmic foundations of reinforcement learning. In a series of papers beginning... 040
Reposted by Levi LelisBen Recht @beenwrekt.bsky.social · 14/02/2025On why it’s time to remove the bias-variance tradeoff from machine learning 101.argmin.netOverfitting to theories of overfittingOn a plot that radicalized me like no other. 5506
Levi Lelis @programsynthesis.bsky.social · 15/02/2025This is one of my favorite types of papers, where strong empirical evidence points in a direction different from where most people had been looking. 010
Reposted by Levi LelisEugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 14/02/2025Model-free deep RL algorithms like NFSP, PSRO, ESCHER, & R-NaD are tailor-made for games with hidden information (e.g. poker). We performed the largest-ever comparison of these algorithms. We find that they do not outperform generic policy gradient methods, such as PPO. arxiv.org/abs/2502.08938 1/N 39321
Reposted by Levi Leliskyunghyuncho.bsky.social @kyunghyuncho.bsky.social · 07/02/2025kyunghyuncho.me/softmax-fore...kyunghyuncho.meSoftmax forever, or why I like softmax – Kyunghyun Cho 35010
Reposted by Levi LelisEugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 06/02/2025We've built a simulated driving agent that we trained on 1.6 billion km of driving with no human data. It is SOTA on every planning benchmark we tried. In self-play, it goes 20 years between collisions. 2229855
Reposted by Levi LelisMarlos C. Machado @marloscmachado.bsky.social · 07/02/2025Martin led this great work, check it out. For a dinosaur like me, let me say that, in more classical RL terms, this is a demonstration of how we can effectively combine options and LLMs through programmatic policies. 091
Reposted by Levi LelisMartin Klissarov @martinklissarov.bsky.social · 04/02/2025Can AI agents adapt zero-shot, to complex multi-step language instructions in open-ended environments? We present MaestroMotif, a method for skill design that produces highly capable and steerable hierarchical agents. Paper: arxiv.org/abs/2412.08542 Code: github.com/mklissa/maestromotif 1216
Levi Lelis @programsynthesis.bsky.social · 04/02/2025I am re-reading "Learning Differentiable Programs with Admissible Neural Heuristics." This is still one of my favorite papers from the past 5–6 years. arxiv.org/pdf/2007.12101 120
Reposted by Levi LelisReinforcement Learning Conference @rl-conference.bsky.social · 29/01/2025RLC call for workshops is out rl-conference.cc/callforworks...! Submissions open on Feb 3rd, deadline on March 7, with the workshops on Aug. 5th. Last year's workshops were inimitable and we @claireve.bsky.social and Josiah Hanna) look forward to your amazing proposalsrl-conference.ccRLJ | RLC Call for Workshops 0114
Reposted by Levi LelisMarlos C. Machado @marloscmachado.bsky.social · 26/01/2025We just released extensive instructions for all the reviewing roles at @rl-conference.bsky.social; ranging from SAC to TR. We are trying something different here that we believe can be better. To ensure we are open about it, we made those instructions public: rl-conference.cc/reviewinstru...rl-conference.ccRLC Review Instructions 2025 0115
Reposted by Levi LelisBailey 🕸️🕷️🦇 @bkacsmar.bsky.social · 09/01/2025Very proud to share that Jialiang Yan from University of Alberta, who I've had the privilege to work with, has been recognized with an honourable mention for CRAs outstanding undergraduate research award!!! www.ualberta.ca/en/computing... 121
Reposted by Levi LelisBailey 🕸️🕷️🦇 @bkacsmar.bsky.social · 08/01/2025Day 1 of our workshop on Private Data Science is underway in sunny San Diego thanks to the encore institute. Co-organized with Rachel Cummings, Amartya Sanyal @amartyasanyal.bsky.social, Clément Canonne @ccanonne.bsky.social and myself, itll be a great three days on privacy! encoredp.github.io 043
Reposted by Levi LelisEugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 07/01/2025Brandon is a wonderful research colleague and I could not endorse enough trying to work with Brandon 1133
Reposted by Levi LelisLoris D'Antoni @lorisdanto.bsky.social · 31/12/2024Three of my (stellar) students are looking for research-oriented internships in PL/FM (but also LLM for code) this coming summer: - Jinwoo Kim (cseweb.ucsd.edu/~jik083/) - Shaan Nagy (cseweb.ucsd.edu/~shnagy/) - Xuanyu Peng (home.dofy.top)cseweb.ucsd.eduJinwoo Kim 251
Levi Lelis @programsynthesis.bsky.social · 31/12/2024This is where programmatic representations can help. In supervised learning, we already have inspiring examples like FlashFill, where a strong inductive bias through a highly constrained hypothesis space allows one to learn complex hypotheses from only 2-3 examples. 150