Alejandro Escontrela

AI Researcher

prof_pic.jpg

Berkeley Way West

2121 Berkeley Way, Berkeley, CA 94704

Hi! I’m Alejandro! I’m currently a 3rd year PhD student at Berkeley AI Research advised by Pieter Abbeel. My work is supported by the NSF and the UC Berkeley Chancellor’s Fellowship.

I am broadly interested in reinforcement learning, unsupervised learning, and robotics. Specifically, my work focuses on extracting behavior priors for RL agents from internet data, making RL super sample efficient, and leveraging large generative models as priors. Before joining BAIR, I completed my undergrad at Georgia Tech, advised by Frank Dellaert. I was also a Research Intern at Google Brain from 2018-2022.

Contact: mail AT escontrela.me

news

Sep 22, 2023 VIPER 🐍 was accepted to NeurIPS in New Orleans! (website , X , code , paper )
Aug 14, 2023 Some of my work was featured in the latest BBC One Panorama documentary . Check out “Beyond Human: Artificial Intelligence and Us”!
May 24, 2023 We released Video Prediction Rewards (VIPER 🐍). Check out our website , Twitter , code , and the paper
Dec 14, 2022 I’m co-organizing the Learning for Agile Robotics workshop at CoRL 2022. Recordings are live on the site!
Oct 22, 2022 Adversarial Motion Priors Make Good Substitutes for Complex Reward Functions was nominated for best paper award at IROS, Kyoto 2022 (11 nominations from >1700 submissions)
Jul 22, 2022 DayDreamer: World Models for Physical Robot Learning was covered by: TechCrunch, MIT Technology Review, Daily Mail, India Times, and more
Jun 30, 2022 Autonomously Untangling Long Cables won Best Systems Paper Award at RSS, NYC 2022

selected publications

  1. Neural Information Processing Systems 2023
  2. Ken Caluwaerts*, Atil Iscen*, J. Chase Kew*, Wenhao Yu*, Tingnan Zhang*, Daniel Freeman, Kuang-Huei Lee, Lisa Lee, Stefano Saliceti, Vincent Zhuang, Nathan Batchelor, Steven Bohez, Federico Casarini, Jose Enrique Chen, Omar Cortes, Erwin Coumans, Adil Dostmohamed, Gabriel Dulac-Arnold,  Alejandro Escontrela
    CoRR 2023
  3. International Conference on Intelligent Robots and Systems 2022
  4. Philipp Wu*,  Alejandro Escontrela*, Danijar Hafner*, Ken Goldberg, and Pieter Abbeel
    Proceedings of Machine Learning Research 2022
  5. Proceedings of Machine Learning Research 2021
  6. Robotics Proceedings 2022