Kevin Lu

Contact: kzl@berkeley.edu
CV / Google Scholar / Twitter / Github

I'm interested in training universal deep learning models. Some of my past work has studied sequence modeling for reinforcement learning and cross-modal transfer. For a recent summary of my work, see the slides for my talk Towards a Universal Paradigm for Decision Making.

I was previously an AI resident at FAIR, advised by Amy Zhang and Yuandong Tian. I graduated with a B.S. in Electrical Engineering and Computer Science from UC Berkeley in 2021, where I did undergraduate research as part of the Robot Learning Lab advised by Igor Mordatch and Pieter Abbeel. Note that I am currently taking a leave from research, and may be slow to answer inquiries.

Highlighted Papers

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
Official: arXiv / website / poster / tweet / code
Press: The Batch article / SyncedReview article / The Gradient article / Yannic Kilcher video / Eindhoven RL seminar

Pretrained Transformers as Universal Computation Engines
Kevin Lu, Aditya Grover, Pieter Abbeel, Igor Mordatch
AAAI Conference on Artificial Intelligence, 2022
Official: arXiv / blog / poster / tweet / code
Press: The Batch article / VentureBeat article / TWIML podcast / Yannic Kilcher video

All Papers

	Pretraining for Language-Conditioned Imitation with Transformers Aaron (Louie) Putterman, Kevin Lu, Igor Mordatch, Pieter Abbeel NeurIPS Offline Reinforcement Learning Workshop, 2021 Official: paper / code
	URLB: Unsupervised Reinforcement Learning Benchmark Michael Laskin, Denis Yarats, Hao Liu, Kimin Lee, Albert Zhan, Kevin Lu, Catherine Cang, Lerrel Pinto, Pieter Abbeel Neural Information Processing Systems (NeurIPS), 2021 Official: arXiv / blog / tweet / code Press: Import AI article
	Reset-Free Lifelong Learning with Skill-Space Planning Kevin Lu, Aditya Grover, Pieter Abbeel, Igor Mordatch International Conference on Learning Representations (ICLR), 2021 Official: arXiv / website / oral / poster / tweet / code
	Efficient Empowerment Estimation for Unsupervised Stabilization Ruihan Zhao, Kevin Lu, Pieter Abbeel, Stas Tiomkin International Conference on Learning Representations (ICLR), 2021 Official: arXiv / website / code
	Adaptive Online Planning for Continual Lifelong Learning Kevin Lu, Igor Mordatch, Pieter Abbeel NeurIPS Deep Reinforcement Learning Workshop, 2019 Official: arXiv / website / oral / poster / code Press: Alignment Newsletter article

Teaching

EECS 126: Probability and Random Processes
Head Teaching Assistant: Spring 2021, Fall 2020
Teaching Assistant: Spring 2020, Fall 2019

CS 70: Discrete Math and Probability
Reader: Spring 2019

Website template from Jon Barron