Alex Lewandowski

Propose some socials for RLC! Research topics, affinity groups, niche interests, whatever comes to mind!

rl-conference.cc/call_for_soc...

I will be presenting a poster at RLDM on Wednesday @ 4:30pm. We show that embedding an agent implicitly constrains both it and the environment. We use this constraint to characterize continual adaptation. lewandowskialex.com/papers/files...

Feel free to reach out if you're attending RLDM!

I will be presenting two posters at ICLR that outlines an optimization perspective on loss of plasticity. Come check them out on Thursday and Friday @ 10am.

Also, feel free to reach out to chat about continual, meta and/or reinforcement learning.

Very cool to see loss of plasticity effects demonstrated at scale.

Congratulations! In my opinion, this level of recognition is over-due. Reinforcement learning is dead, long live reinforcement learning!

The whole magic of Bayes is that it is computationally unconstrained (e.g., AIXI). I would go so far as to say that computationally constrained Bayes is the more fundamental theory, and probably doesn't look like Bayes anymore.

I would be interested in seeing papers on this topic though!

We 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...

The supervised learning dataset is basically a simulator.

Scaling for the people, by the people! Hear, hear!

(Of course, benchmarks are always a little silly)

One contributing factor is our uncertainty about the potential capabilities of AI.

Without hard limits on AI, people's pre-existing worldviews influence whether they imagine the best or worst case scenarios.

Malazan: It's long, and a love it or hate it kind-of series. Still, worth trying.

And please include the Arxiv link on the first post in a thread, and not the last one.

Too bad the reviewers and authors can't upload "alpha-level simulations" of themselves so that the bots can sort it out on openreview instead of all this point-courting.

Picture of Trinity College Dublin's campanile in the front square of the college. Writing overlayed to indicate that the RLDM conference will take place from the 11th to the 14th of June, 2025. Logos for Trinity College and the RLDM conference at the bottom of the image.

Come join us at Trinity College Dublin for the 6th Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM2025)

Abstract deadline January 15: rldm.org/submit

reinforcement learning