Sebastian Curi

Ph.D. Student in CS @ETH.

[Sebastian Curi @sebascuri]

@lucasllach Simplemente la ley del offside tiene que ser al “reves”: si cualquier parte del defensor esta habilitando al delantero entonces es valida la jugada.

[Giorgia Ramponi @ICML @gio_ramponi]

I’m looking for one strong RL intern (4 months, starting in September) with the intention of becoming a PhD position. The position is based in Zurich, a fantastic research environment with an active ML/AI community.

info here: https://t.co/XXBUPvpocJ

[Sebastian Curi @sebascuri]

@SusanPolgar Nc7+ Kb8, Nxd5+ Ka8, Nb6+ Nxb6, Rxa7+ Kxa7, Ra5#

[Sebastian Curi @sebascuri]

@arkrause @UN Congrats Andreas!

[Sebastian Curi @sebascuri]

@davlindner Congrats!!

[Sebastian Curi @sebascuri]

@GMJacobAagaard Ra5

[Sebastian Curi @sebascuri]

@TheZugzwangBlog Qf6

[Sebastian Curi @sebascuri]

Great Work by David!

[David Lindner @davlindner]

Our #ICML2022 paper argues that representing learned human preferences as constraints might be a more robust alternative to only representing them as reward functions. This motivates looking at actively learning about unknown constraints from (expensive) human feedback. 🧵👇

[Mojmir Mutny @mutny_ml]

Done with NeurIPS? Submission deadline is in 9 days for the ReALML @ #icml2022 workshop (https://t.co/0z7pp6xEbv)!

New to the field? Experimental design is a classic field spanning many decades—check out this video from the 1960s for some history: https://t.co/fFrpSsXBci

[Lenore Blum @BlumLenore]

Dear @pmddomingos, Not sure what your problem is. But I take your insults personally. I’d appreciate your affording me the courtesy of reading my story below, and checking out the included links. This story is not mine alone. Sincerely, @blumlenore 1/7

[Camilo Toledo @bsas_lovers]

Atardecer en Buenos Aires. Dale retuit si te gustó así me ayudás a compartir mi trabajo fotográfico!😉
Av. Corrientes, Buenos Aires, Argentina.

[Ilija Bogunovic @ilijabogunovic]

Check out the LAS blog and new posts on:
- Towards Bayesian Causal Structure Learning in Real-World Applications
- How to explore to find a robust control policy
- Sample efficient reward learning for RL
Details 👉 https://t.co/0yQl2z6k1I

[Sebastian Curi @sebascuri]

@vincefort @gxr @arkrause @s_mandt @wellingmax Congrats

[Ilija Bogunovic @ilijabogunovic]

Looking to hire a Ph.D. student to work on algorithmic data-driven decision making (Topics: Multi-agent learning, RL, Bayesian optimization, Experiment design) at UCL E&EE @ucleeenews. Retweets appreciated!

Details: https://t.co/gkVdsB5ymJ

[Sebastian Curi @sebascuri]

If you reached this point, I hope to see you today at the virtual Poster Session 3, on spot A5!

[Sebastian Curi @sebascuri]

Welcome to my first thread on my work! Today at 5 pm CET I will be presenting at @icmlconf our work on Robust RL with @ilijabogunovic and @arkrause.
Paper: https://t.co/ZlyqtJzrNH
Blog: https://t.co/RGX1q1xR3j
Code: https://t.co/hKNrPYJ1Y4

[Sebastian Curi @sebascuri]

In a suite of Mujoco Environments, RH-UCRL has better robust performance than different benchmarks and ablations, and it also has a good average performance.