Aravind Rajeswaran

Research Scientist at Meta AI (FAIR)

Visiting PostDoc / Collaborator,
Berkeley AI Research Lab (BAIR)

I'm interested in building generalist AI agents that can operate in the open world. Towards this goal, my research brings together elements of reinforcement learning, representation learning, and world models. In addition to my role at FAIR, I also collaborate with the academic research labs of Prof. Pieter Abbeel at UC Berkeley and Prof. Abhinav Gupta at CMU. I recieved my PhD in Computer Science from the University of Washington working with Profs. Sham Kakade and Emo Todorov. During this time, I also worked closely with Sergey Levine and Chelsea Finn, and spent time as a student researcher at Google Brain and OpenAI. Before that, I recieved my bachelors and the best undergraduate thesis award from IIT Madras working with Balaraman Ravindran.

Representative Projects

Foundation Models and SSL for Agents

OpenEQA: Embodied Question Answering in the Era of Foundation Models

Cortex Team @ FAIR (Role: PI and Project Lead)

Computer Vision and Pattern Recognition (CVPR) 2024 | Project Website

R3M: A Universal Visual Representation for Robot Manipulation

Suraj Nair, Aravind Rajeswaran, Vikash Kumar, Chelsea Finn, Abhinav Gupta

Scaling Robot Learning Workshop at ICRA 2022 | Best Paper Award | Project Website

VC-1: Where are we in the search for an Artificial Visual Cortex for Embodied Intelligence?

Cortex Team @ FAIR (Role: co-PI and Tech Lead)

Neural Information Processing Systems (NeurIPS) 2023 | Project Website

Reinforcement Learning

Decision Transformer: Reinforcement Learning via Sequence Modeling

Lili Chen*, Kevin Lu*, Aravind Rajeswaran, Kimin Lee, Aditya Grover, Michael Laskin, Pieter Abbeel, Aravind Srinivas, Igor Mordatch

Neural Information Processing Systems (NeurIPS) 2021 | Project Website

Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrations

Aravind Rajeswaran, Vikash Kumar, Abhishek Gupta, John Schulman, Emanuel Todorov, Sergey Levine

Robotics: Science and Systems (RSS) 2018 | Project Website

Masked Trajectory Models for Prediction, Representation, and Control

Philipp Wu, Arjun Majumdar, Kevin Stone, Yixin Lin, Igor Mordatch, Pieter Abbeel, Aravind Rajeswaran

International Conference on Machine Learning (ICML) 2023 | Project Website

Mentoring

I enjoy collaborating with a diverse set of students and researchers. I have had the pleasure of mentoring some highly motivated students at both the undergraduate and PhD levels.

Interns

Philipp Wu (PhD at Berkeley)
Arjun Majumdar (PhD at Georgia Tech)
Anurag Ajay (PhD at MIT)
Mandi Zhao (PhD at Columbia University)
Nicklas Hansen (PhD at UC San Diego)
Shikhar Bahl (PhD at CMU)
Allan Zhou (PhD at Stanford)
Suraj Nair (PhD at Stanford)
Liyiming Ke (PhD at UW Seattle)
Andi Peng (PhD at MIT)
Yuchen Cui (PhD at UT Austin)

University Students and AI Residents

Aryan Jain (UC Berkeley, BS)
Sergio Arnaud (AI Resident, FAIR)
Rafael Rafailov (Stanford, MS ==> Stanford, PhD)
Karmesh Yadav (CMU MS ==> Georgia Tech, PhD)
Kevin Lu (UC Berkeley, BS ==> FAIR AI resident)
Ben Evans (UW, BS/MS ==> PhD at NYU)
Sarvjeet Ghotra (IITM Intern ==> PhD at MILA)
Colin Summers (UW, BS/MS ==> UW PhD)
Catherine Cang (UC Berkeley, BS ==> SWE at Plaid)
Divye P. Jain (UW, BS/MS ==> Google MTV)

Teaching

CSE599G: Deep Reinforcement Learning (Instructor)

I designed and co-taught a course on deep reinforcement learning at UW in Spring 2018. The course presents a rigorous mathematical treatment of various RL algorithms along with illustrative applications in robotics. Deep RL courses at UW, MIT, and CMU have borrowed and built upon the material I developed for this course.

CSE547: Machine Learning for Big Data (Teaching Assistant)

This is an advanced graduate level course on machine learning with emphasis on machine learning at scale and distributed algorithms. Topics covered include hashing, sketching, streaming, large-scale distributed optimization, federated learning, and contextual bandits. I was the lead TA for this class.