don’t be misled by low-order statistics

I remember when I was doing my phd I always enjoyed going to conferences since when accepted I got a 20 min slot to talk about my work and have discussions. The field grew and this is over now. Most papers get a poster slot where discussions happen. That is ok but I think there

I am unhappy with the conference posters and decided to host a series of presentations with excellent young and senior researchers who could teach us about the fantastic stuff they have been doing. I hope this aids in making research more accessible! ♿️ I title the series Mate

I am very excited that we have uploaded the 2nd lecture in the introduction to #RL series on the channel. In this talk, we finish the example from the last lecture: youtube.com/watch?v=SpbkQ1… and dive into value functions: youtu.be/zSenci12ayI. We provide example computations

From Paris to Warsaw and Edinburgh: Explore the latest job opportunities in our European #AI network! ➡️ ellis.eu/jobs #machinelearning #artificialintelligence #jobs

Our Multi-Agent RL framework EPyMARL, which extends the original PyMARL framework of WhiRL, is widely used in MARL research (nearly 400 stars on Github!) and several MARL benchmarks have been built on it. We list three in this thread! Blog: agents.inf.ed.ac.uk/blog/epymarl/ (1/4)

IMP-MARL: a Suite of Environments for Large-scale Infrastructure Management Planning via MARL by Pascal Leroy, Pablo G Morato et al. NeurIPS 2023 Paper: arxiv.org/abs/2306.11551 (3/4)

Want to do a PhD in RL/MARL in our group at School of Informatics, The University of Edinburgh? Our brand new 4-year PhD program in Sensing, Processing, and AI for Defence and Security (SPADS) is now open for applications! Deadline April 10, 2024. Contact me if interested. spads.ac.uk

New blog post: Our Policy for Reproducibility & Transparency in Research Includes action checklist that people can use in their research. ☑️ agents.inf.ed.ac.uk/blog/reproduci…

Self-driving! Multi-agent learning! RL on 100 deployed cars! I will tell you about all of these things

Looking forward to presenting at the University of Hong Kong next week! My talk is on June 21st 2-4pm (Beijing time). See flyer below for talk details ⬇️

BREAKING NEWS The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Physics to John J. Hopfield and Geoffrey E. Hinton “for foundational discoveries and inventions that enable machine learning with artificial neural networks.”

Eugene is doing fantastic research in decision making for real-world multi-agent problems. Consider applying!

Becoming a better researcher is easy. Just spend your time like so: 40% reading papers 40% writing code 40% coming up with new ideas 40% studying textbooks 40% talking to other researchers

I am happy to announce that the first draft of my RL tutorial is now available. arxiv.org/abs/2412.05265

If you were at the last RLC NeurIPS event, you know it's not to be missed. Invite any RL researcher you know, no invite-only parties here.

Out NOW - Our Multi-Agent RL textbook has finally arrived in stores today! Order from The MIT Press @mitpress.bsky.social or other book stores. Just in time for Christmas! 🙂 Lecture slides (with Tex files) and algorithm code are available from the book homepage. See Lukas' post ⬇️ for more details.

When I was an undergrad, one of the biggest hurdles preventing me from exploring deep learning was limited GPU resources. Even today in academia, seeing industry teams use thousands of GPUs for LLM training can make hands-on experimentation feel completely out of reach. Tools