Missed our workshop at @rl-conference.bsky.social?
Watch all keynote presentations and the panel discussion on YouTube:
youtube.com/playlist?lis...
Big thanks to our awesome speakers: @togelius.bsky.social, @jmac-ai.bsky.social, Roberta Raileanu, Michael Bowling and @pcastr.bsky.social!
Conference rejections: hurts a bit but it's fine.
Radiohead rejections: I'll be crying in a corner all night
This one's been a long time coming.
In this post on Decisions & Dragons I answer "Should we abandon RL?"
The answer is obviously no, but people ask because they have a fundamental misunderstanding of what RL is.
RL is a problem, not an approach.
www.decisionsanddragons.com/posts/should...
The @rl-conference.bsky.social is always the best time of the year for a RL researcher!
This year I had a blast co-organizing the @rlvgworkshop.bsky.social, alongside amazing people! A huge thanks to our keynote speakers and everyone who joined the discussion.
Canโt wait for Montreal next year!
Weโre excited to announce our final keynote speaker: @michaelhbowling.bsky.social (University of Alberta). ๐ฒ๐
In this talk, he will explore what drew him to explore using 40-year-old video games to test RL agents!
Join us at @rl-conference.bsky.social to find out more about the Atari Benchmark!
Weโre excited to announce our next speaker: @pcastr.bsky.social (Google DeepMind, Mila) ๐ฎ๐น๏ธ
His research tackles core challenges in deep RL for real-world impact, like network scalability and plasticity loss!
Join us at @rl-conference.bsky.social to learn more about Pabloโs work!
Weโre excited to announce our next keynote speaker:
@neuroai.bsky.social (Microsoft Research) ๐ง
Ida studies how humans and AI build world models for memory, exploration, and planning. Her talk will show how game-playing AI reveals insights into human and artificial cognition.
Join us at #RLC2025!
Weโre excited to announce our next speaker: Roberta Raileanu ( Google DeepMind)!
Roberta will discuss the history and future of NetHack, the iconic roguelike that continues to challenge RL and LLM agents alike. โ๏ธ
Join us to learn how we can develop agents capable of tackling open-endedness!
Weโre excited to announce our first speaker: James MacGlashan (@jmac-ai.bsky.social), from Sony AI! ๐
James led the development of GT Sophy, a superhuman RL racing agent in Gran Turismo!
Join his talk to learn about how GT Sophy was used as an opponent in GT7 and whatโs next for RL in gaming.
Two days left to submit to our Reinforcement Learning and Video Game workshop at @rl-conference.bsky.social !
Join us in discussing the bright future of RL and Video Games!
โA delayed paper is eventually good; a bad paper is forever badโ
- Shigeru Miyamoto
One more week to submit to our amazing RL and video games workshop!
โผ๏ธ Two days left to submit your contributions for our workshop!
Working on RL and video games? Join us! We will have amazing speakers, cool events and a ๐ฎspecial topical๐ฎ best paper award prize!
โ Paper Submission Deadline: May 30th
Find out more here: sites.google.com/view/rlvg-wo...
Fantastic work on general utility MDPs by @pedrosantospps.bsky.social!
You can find a summary of the paper in his account (go follow him as well!).
One week left to submit your contributions for our workshop! โฐ๐ฎ
We welcome submissions on video games to benchmark open problems in RL and in applications of RL for video game development!
โ Paper Submission Deadline: May 30th
Find out more here: sites.google.com/view/rlvg-wo...
Running out of time to submit your contributions for our workshop!
We welcome submissions on video games to benchmark open problems in RL and in applications of RL for video game development!
โ Paper Submission Deadline: May 30th
Find out more here: sites.google.com/view/rlvg-wo...
Happy to share that our paper "The Number of Trials Matters in Infinite-Horizon General-Utility Markov Decision Processes" got accepted as a spotlight poster at the International Conference on Machine Learning (ICML).
๐Excited to share that our paper "๐๐ฎ๐ฆ๐๐ง-๐๐ฅ๐ข๐ ๐ง๐๐ ๐๐ฆ๐๐ ๐ ๐๐จ๐๐๐ฅ๐ฌ ๐๐ฆ๐ฉ๐ซ๐จ๐ฏ๐ ๐๐ข๐ฌ๐ฎ๐๐ฅ ๐๐๐๐จ๐๐ข๐ง๐ ๐๐ซ๐จ๐ฆ ๐ญ๐ก๐ ๐๐ซ๐๐ข๐ง" has been accepted to #ICML2025!
๐ Preprint Link: arxiv.org/abs/2502.03081
Huge thanks to my co-authors!
This year I am co-organizing the 8th iteration of the Tractable Probabilistic Modeling #TPM workshop at #UAI2025 ๐ด Rio de Janeiro edition ๐ด
๐ lnkd.in/dDK8T5Au
โฐ Submission deadline: May 23th AoE
๐ด Date: July 15th
๐งต๐
Join us at the Reinforcement Learning Conference 2025 for a Workshop on Reinforcement Learning and Video Games, celebrating a decade since the Atari Nature paper ๐ฎ ๐น๏ธ
Check out our webpage for more information: sites.google.com/view/rlvg-wo...
Itโs been 10 years since the Atari DQN Nature paper! What better way to celebrate this milestone than with a dedicated workshop on RL and Video Games at @rl-conference.bsky.social? ๐ฎ๐ฎ
Join us to discuss how video games can be used to address fundamental RL problems and how RL can be used in Games!
The reply-all button should be banned.
Especially when organization-wide mailing lists are in CC.
Rant over.
The acknowledgement check for ICML reviews is a bit like global warming.
We see it, we promise we will change, but, in the end, nothing happens.
So happy to finally see @kthuniversity.bsky.social in Bluesky!
Hope more European universities follow suit and expand their academic and research presence here.
Today is a historical day for the RL community (another one!).
Congratulations to Andrew and Rich! Well deserved!
awards.acm.org/about/2024-t...
RLC Keynote speakers: Leslie Kaelbling, Peter Dayan, Rich Sutton, Dale Schuurmans, Joelle Pineau, Michael Littman
Some extra motivation for those of you in RLC deadline mode: our line-up of keynote speakers -- as all accepted papers get a talk, they may attend yours!
@rl-conference.bsky.social
Iโm attending #NeurIPS in Vancouver from Tuesday to Sunday.
Iโll be presenting our paper โAmplified Early Stopping Bias: Overestimated Performance with Deep Learningโ in the Scientific Methods for Understanding Deep Learning Workshop.
Letโs meet especially if you are interested in Neuroscience!
This work wouldnโt be possible without our fantastic collaborators:
- Francisco S. Melo (Instituto Superior Tรฉcnico, Lisbon)
- Antรดnio H. Ribeiro (Uppsala University)
- Marten Bjรถrkman (KTH)
- Danica Kragic (KTH), whoโs also giving a keynote at #NeurIPS2024!
Nona (@nonar.bsky.social) will present our workshop paper "Amplified Early Stopping Bias: Overestimated Performance with Deep Learning". A critical look at early stopping in low-data regimes.
Thread: bsky.app/profile/nona...
Poster: neurips.cc/virtual/2024...
Paper: openreview.net/forum?id=wbU...
Bernardo (@esteveste.bsky.social) will present NeuralSolver, our new approach to learn decision-making algorithms that scale to large inputs at test time.
Thread: bsky.app/profile/este...
Poster: neurips.cc/virtual/2024...
OpenReview: openreview.net/forum?id=IxR...
Farzaneh (@farzaneh-taleb.bsky.social) will present our spotlight paper "Can Transformers Smell Like Humans?". We explore the alignment between human olfactory perception and foundation model representations.
Poster: neurips.cc/virtual/2024...
Paper: openreview.net/forum?id=3f8...
