Joined October 2020
9 Photos and videos
RL for real-world applications = offline learning reward learning. How do we make this work? Find out more at ICLR poster #377 at 10am today! @gio_ramponi and I will be presenting our latest work on offline preference-based RL (joint w/ @gxr, @bschoelkopf).
2
7
52
7,905
Alizée Pace retweeted
I am at ICLR this week ! Excited to be presenting our work on identifiability with @rpatrik96 @JuhosAttila @randall_balestr @klindt_david @wielandbr. Tune in on Saturday at 11AM in Garnet 216-218 for our oral presentation and at 3PM in Hall 3/2B for our poster!
1
12
86
6,495
Alizée Pace retweeted
🎉 Yesterday, @AlizeePace, our very first PhD Fellow of the @ETH_AI_Center, graduated! She was supervised by ETH AI Center Faculty members Prof. Gunnar Rätsch @gxr and Prof. @bschoelkopf. Congrats, Dr. Pace! Next, she will join @GoogleDeepMind in Zurich as a Research Scientist.
6
2
67
4,925
Last Friday, I had the pleasure of giving an invited talk at the workshop on Reinforcement Learning in Mannheim, Germany! I presented recent work on offline preference-based reinforcement learning, aiming to make RL more practical for real-world applications like healthcare.
1
2
37
5,077
This is joint work with @bschoelkopf, @gxr, and @gio_ramponi, which we will also be presenting at ICLR 2025 🎉 Link: arxiv.org/abs/2406.18450

1
1
6
1,020
Very grateful to the organisers @vernadec, Leif Döring, and Simon Weißmann for inviting me and putting together a fantastic event.
3
437
Alizée Pace retweeted
✨New Preprint ✨ Ever thought that reconstructing masked pixels for image representation learning seems sub-optimal? In our new preprint, we show how masking principal components—rather than raw pixel patches— improves Masked Image Modelling (MIM). Find out more below 🧵
17
61
524
48,467
Alizée Pace retweeted
🚨 Recruiting PhD students to join my team at @DTU_VisualComp & @AiCentreDK on 3D Vision! 🇩🇰 🚲🌊🏰 Copenhagen is an emerging hub for computer vision with a thriving community—and it’s an amazing city! 📢 Apply now: efzu.fa.em2.oraclecloud.com/… & please reach out with any questions!

1
17
37
5,227
Alizée Pace retweeted
Machine learning has made incredible breakthroughs, but our theoretical understanding lags behind. We take a step towards unravelling its mystery by explaining why the phenomenon of disentanglement arises in generative latent variable models. Blog post: carl-allen.github.io/theory/…
2
12
56
7,118
Alizée Pace retweeted
🔬 Advance the frontiers of AI: @ETH_AI_Center Fellowship Programs –#PhD & #Postdoc Opportunities 🔬 💫Push the boundaries of Reinforcement Learning and Data-driven Control💫 ✍️ Apply by November 19, 2024: ttps://ai.ethz.ch/apply
1
5
9
2,089
Alizée Pace retweeted
Currently hiring research interns. If you're interested in contributing to RL research write me!
43
148
1,075
191,734
Alizée Pace retweeted
Had a blast at the #ICML2024 workshops @arlet_workshop and FoRLaC! It was really cool to see everyone’s enthusiasm and organizing ARLET was a lot of fun. Thanks to everyone involved for making it such an enjoyable experience, see you at the next edition! 🫡
2
3
45
9,033
Excited to present recent work on how to efficiently query preference feedback for offline RL: arxiv.org/abs/2406.18450 @mhf_icml2024 yesterday, now see you at the FoRLaC workshop at 10:30! Joint work with @gio_ramponi @gxr @bschoelkopf

2
5
44
6,470
Alizée Pace retweeted
It‘s brilliant to see generations of my students meet at conferences like @icmlconf! Here my former master student @AlizeePace with my current PhD student Kasia Kobalczyk after a great discussion about Kasia‘s fantastic paper on informed Meta-Learning: arxiv.org/abs/2402.16105
1
2
17
3,140
Alizée Pace retweeted
26 Jul 2024
The panel discussion is starting!!
1
2
13
1,375
Alizée Pace retweeted
Replying to @arlet_workshop
@arlet_workshop started now! Come to the first floor to join us #ICML2024
4
12
1,435
Don’t miss our workshop at the intersection of RL theory & practice tomorrow!
Replying to @arlet_workshop
@arlet_workshop is tomorrow at #ICML2024 (Schubert 1-3)! We'll discuss the gap between theoretical and experimental developments in #RL with an amazing line-up of speakers/panelists:@white_martha,Dylan Foster, Niao He, @svlevine, @j_foerst, Akshay Krishnamurthy, Marcello Restelli
8
1,636
Alizée Pace retweeted
We are delighted to introduce🕴️J-BOND🕴️, a novel RLHF algorithm to align LLMs via Best-of-N Distillation: arxiv.org/abs/2407.14622

5
17
80
14,271
Thrilled to be returning to @GoogleAI for a research internship! I’ll be again part of the Gemini team, developing RLHF and reward modelling approaches to improve model quality. Exciting few months ahead!
1
2
82
18,866