Lior Shani

We're excited to share our new paper: "Personalized and Sequential Text-to-Image Generation"! Check out the paper and our new sequential human rater dataset! 👇 Paper: arxiv.org/pdf/2412.10419 Dataset: kaggle.com/datasets/googl… Details below.. 1/N 🧵

Check out our paper for more details on how we're pushing the boundaries of RLHF for more natural and effective multi-turn conversations! arxiv.org/abs/2405.14655 @LiorShan @aviv_rosenberg @AsafCassel (8/8)

Experimental results showing MTPO outperforms single-turn RLHF baselines and a multi-turn generalization of RLHF. This demonstrates the effectiveness of our approach in improving the quality of multi-turn conversations. (7/8)

Excited to share our latest research on aligning Large Language Models (LLMs) with human preferences! We're moving beyond single-turn interactions to improve multi-turn conversations. arxiv.org/abs/2405.14655 Joint work by @GoogleResearch @GoogleDeepMind (1/8)

Check out our paper here: arxiv.org/pdf/2205.15376… Code can be found here: github.com/guytenn/Termin… And of course, thank you to everyone who collaborated with me on this project: @NadavMerlis, @LiorShan, @shiemannor, @ShalitUri, @GalChechik, Assaf Hallak, and @DalalGal. (9/n) n=9

Finally, we discuss the need for exploration in imitation learning, and argue that our Apprenticeship Learning based approach which relies on the MDP structure, is superior to supervised learning approaches such as BC.

We show our approach is both theoretically efficient and practical: we provide regret guarantees and show how to avoid solving an MDP at each iteration as in prior works. This allows us to devise a well-performing deep RL implementation of our (OAL) algorithm.

Glad to present our paper "Online Apprenticeship Learning" at #AAAI2022 arxiv.org/abs/2102.06924 with @TZahavy @MannorShie We show how to efficiently reproduce experts' behavior from an offline data of trajectories, by interacting with the MDP (when rewards are not specified).

teleporting, swimming with sharks, and lots more. It turned out better than I've ever expected! Now, before releasing it out to the world, I'm looking for a partner that can help me market the game properly. You can help me out by retweeting! Promo:

Ludwig @Princeton Director Joshua Rabinowitz, a pioneer of metabolomics, has contributed to the development of a cancer therapy & undone enduring assumptions about metabolism. His work is opening new approaches to cancer therapy. Learn more: bit.ly/3Ex0ndx

Come chat today at the deep RL workshop @NeurIPSConf!

Applying Mirror Descent in deep RL is nice and easy! slideslive.com/38941342/mirro…

@manan_tomar slideslive.com/38941342/mirro…

Overall, MDPO is an easily scalable policy optimization algorithm with minimal hyper-params/heuristics involved, and is nicely grounded in mirror descent theory :) Joint work with @LiorShan, Yonathan Efroni, Mohammad Ghavamzadeh Come chat on Dec 11, 11:30 am PST!

Mirror Descent has been utilized in RL theory quite extensively, but can we build practical RL algorithms from it? Our contributed talk at the deep RL workshop @NeurIPSConf discusses exactly this! Paper: arxiv.org/abs/2005.09814 Code: github.com/manantomar/Mir… Details below!

Prof. Shie Mannor is presenting our work at the great RL theory seminar this Tuesday! The talk will be about the connections between TRPO and convex optimization, possible practical implications and how to explore in policy optimization...

Prof. Shie Mannor is presenting our work at the great RL theory seminar this Tuesday! The talk will be about the connections between TRPO and convex optimization, possible practical implications and how to explore in policy optimization...