Yevgen Chebotar (@YevgenChebotar) / X

Yevgen Chebotar

45 posts

Yevgen Chebotar

@YevgenChebotar

Robotic foundation models @NVIDIA 🤖 Previously @GoogleDeepMind (RT-2, VLAs, Offline RL) and @Figure_robot (Helix)

Joined March 2017

Yevgen Chebotar
@YevgenChebotar
Sep 7, 2023
Offline RL strikes back! In our new Q-Transformer paper, we introduce a scalable framework for offline reinforcement learning using Transformers and autoregressive Q-Learning to learn from mixed-quality datasets! Website and paper: q-transformer.github.io 🧵
GIF
211K
Yevgen Chebotar
@YevgenChebotar
Mar 11, 2024
Some personal updates! Excited to join the team @Figure_robot to help building AI for the robot age! 🤖
66K
Yevgen Chebotar
@YevgenChebotar
Jun 14, 2021
Excited to present our work on Actionable Models at #ICML! Find the camera-ready version at arxiv.org/abs/2104.07749 In this work, we learn functional understanding of the world through goal-conditioned Q-learning and use it for reaching visual goals or learning downstream tasks.
Yevgen Chebotar
@YevgenChebotar
Jul 28, 2023
Excited to present RT-2, a large unified Vision-Language-Action model! By converting robot actions to strings, we can directly train large visual-language models to output actions while retaining their web-scale knowledge and generalization capabilities! robotics-transformer2.github.io
00:26
Google DeepMind
@GoogleDeepMind
Jul 28, 2023
Today, we announced 𝗥𝗧-𝟮: a first of its kind vision-language-action model to control robots. 🤖 It learns from both web and robotics data and translates this knowledge into generalised instructions. Find out more: dpmd.ai/introducing-rt2
24K
Yevgen Chebotar
@YevgenChebotar
Apr 19, 2021
Excited to present our new work on Actionable Models, an approach for learning functional understanding of the world via goal-conditioned Q-functions in a fully-offline setting! paper: arxiv.org/abs/2104.07749 website: actionable-models.github.io youtube.com/watch?v=S3SCR7…
arxiv.org
Actionable Models: Unsupervised Offline Reinforcement Learning of...
We consider the problem of learning useful robotic skills from previously collected offline data without access to manually specified rewards or additional online exploration, a setting that is...
Yevgen Chebotar
@YevgenChebotar
May 19, 2019
Excited to present our work on closing the sim-to-real loop at ICRA in Montreal! Please visit our poster and talk on Tuesday! “Closing the Sim-to-Real Loop: Adapting Simulation Randomization with Real World Experience” Paper: arxiv.org/abs/1810.05687 Video: youtube.com/watch?v=nilcJY…
Yevgen Chebotar
@YevgenChebotar
Mar 6, 2024
RT-H learns a hierarchy all the way from high-level tasks through low-level “language motions” to robot actions! ✅ Improved performance and generalization through better data sharing ✅ Automated grounded “bottom-up” labeling ✅ Ability to intervene and correct with language
Suneel Belkhale
@suneel_belkhale
Mar 6, 2024
Is language capable of representing low-level *motions* of a robot? RT-Hierarchy learns an action hierarchy using motions described in language, like “move arm forward” or “close gripper” to improve policy learning. 📜: arxiv.org/abs/2403.01823 🏠: rt-hierarchy.github.io (1/10)
00:00
4.6K
Yevgen Chebotar
@YevgenChebotar
Oct 16, 2018
Improve the simulation to reality robotic skill transfer by closing the sim-to-real loop and adjusting simulation randomization! Paper: arxiv.org/abs/1810.05687 youtu.be/nilcJY5Kdt8
Yevgen Chebotar
@YevgenChebotar
Jun 14, 2019
Our new work on performing meta-learning using learned loss functions! Also visit our talk and poster at Multi-Task and Lifelong Reinforcement Learning Workshop at ICML tomorrow! Paper: arxiv.org/abs/1906.05374 w/ @amolchanov86, S. Bechtle, @ludo_righetti, @_kainoa_, G. Sukhatme
Yevgen Chebotar
@YevgenChebotar
Nov 9, 2023
Presenting RT-2 poster at CoRL!
Quan Vuong
@QuanVng
Nov 9, 2023
Pictures taken at RT-2 poster at @DannyDriess requests ; ) @YevgenChebotar We miss you @TianheYu CC @hausman_k
RT-2: Vision-Language-Action Models
From robotics-transformer2.github.io
3.8K
Yevgen Chebotar
@YevgenChebotar
May 16, 2024
Congrats everyone, 170+ authors and contributors, great to see the robotic field coming together!
Karl Pertsch
@KarlPertsch
May 16, 2024
Our OpenX paper won best paper at ICRA! Congrats to all my co-authors! 🎉🎉 This is an ongoing effort, we recently added new datasets from the community that double the size of the OpenX dataset -- keep 'em coming! :) Check datasets & how to contribute: robotics-transformer-x.github.io
2.8K
Yevgen Chebotar
@YevgenChebotar
Sep 7, 2023
Replying to @YevgenChebotar
By using autoregressive Bellman updates, conservative regularization, Monte Carlo and n-step returns, we are able to combine human demonstrations and autonomously collected data to learn multi-task language-conditioned policies from both, successful and failed examples.
2.5K
Yevgen Chebotar
@YevgenChebotar
Sep 7, 2023
Replying to @YevgenChebotar
Our real robot policies significantly improve upon RT-1 and other baselines when trained on limited amount of human demonstrations by leveraging autonomously collected negatives and dynamic programming properties of Q-learning.
00:00
2.1K
Yevgen Chebotar
@YevgenChebotar
Mar 7, 2024
Turns out classification loss works surprisingly well for value-based RL, also some nice gains when used with Q-Transfomer!
Aviral Kumar
@aviral_kumar2
Mar 7, 2024
Super simple code change to get value-based deep RL scale *much* better w/ big models across the board on Atari games, robotic manipulation w/ transformers, LLM + text games, & even Chess! Just use classification loss (i.e., cross entropy), not MSE!! arxiv.org/abs/2403.03950🧵⬇️
5.8K