R. Alessio @ BU

Going to #ICML2025 next week, ping me up if you are interested in pure exploration problems in #RL! Also check my poster on Exploration for Policy Evaluation, Tuesday morning 11.30am! Joint work with @aldopacchiano at BU.

@NeurIPSConf, why take the option to provide figures in the rebuttals away from the authors during the rebuttal period? Grounding the discussion in hard evidential data (like plots) makes resolving disagreements much easier for both the authors and the reviewers. Left: NeurIPS…

As @roydanroy & others noted, review quality has degraded, potentially AI-generated. Isn't it a viable next step to have a well-trained LLM reviewer perform a first round of review to reduce the load? For quality reviews in 2nd round? #NeurIPS2025 #icml2025 #ACL2025 #iclr2025

Had an amazing session at #ICML2025! Happy to have presented our last paper on exploration for policy evaluation in #RL, a joint work with @aldopacchiano at #BU Check the paper here arxiv.org/abs/2502.02516 And follow us for more work on principled exploration in RL!

Three works accepted to ICML: 1. Multiple-policy Evaluation via Density Estimation - led by PhD student Yilei Chen, joint with Ioannis Paschalidis. 2. Adaptive Exploration for Multi-Reward Multi-Policy Evaluation - led by Alessio Russo. (1/2)

Our work "Pure Exploration with Feedback Graphs" is being presented by Alessio @rssalessio at AISTATS 2025!

Interested in #RL, #bandits, and #learning in general? This is a collection of interesting papers (and books) that I have read so far or want to read. Note that the list is not up-to-date. github.com/rssalessio/rea…

Not really happy with the #ICML review process this year. More than 10 forms, and one does not know where to write what.

Sounds more like a warning than a gentle reminder. #icml

Hai buttato 150 miliardi di € (e il conto finale ha ancora da venire) per rifare le villette, senza dare un euro a scuola, sanità e welfare, e adesso sbraiti che si vogliono investire soldi per la difesa anziché darli a scuola, sanità e welfare? Ma non ti vergogni?

I still see papers in #adversarial #RL trying to study stealthy attacks, but almost none of them consider that perturbations in one state affect future states. A definition of stealthy attack should account for the dynamics of the MDP too. Check also arxiv.org/pdf/2109.07171

Interested in #exploration in #RL and currently at #NeurIPS2024 ? Come by our poster "Multi-Reward Best Policy Identification" We tackle the exploration problem for multiple rewards in tabular and #deeprl Tomorrow afternoon (Wed), West Ballroom A-D #6608