I'm an incoming PhD student at Berkeley advised by Sergey Levine. Previously I had the privilege to work with Abhishek Gupta in WEIRD Lab at the University of Washington during my undergrad and masters.
I am currently interning at the Bosch Center for AI in Pittsburgh working on adaptation and world modeling. I have previously interned at the Amazon Robotics Innovation Lab on a prototype package-manipulation robot called Calisto.
I want to train robots that are robust and capable enough to deploy in real environments. To this end, I'm currently excited about multiple directions, including: leveraging off domain data and pretrained priors, data curation, world modeling, and test time adaptation.
Outside research, I love rock climbing, my dog, travel, and skiing! I also care a lot about teaching and have been an instructor or TA for nearly all of my undergraduate and masters years.
I'm interested in robotics, reinforcement learning, and world modeling.
Jacob Berg, Chuning Zhu, Yanda Bao, Ishan Durugkar, Abhishek Gupta
Semantic World Models, unlike traditional world models, answer questions about the future given current observations and a sequence of actions.
Marius Memmel*, Jacob Berg*, Bingqing Chen, Abhishek Gupta†, Jonathan Francis†
STRAP uses dynamic time warping and pretrained visual foundation models to retrieve sub-trajectories that augment policy learning.
Jianlan Luo, Zheyuan Hu, Charles Xu, You Liang Tan, Jacob Berg, Archit Sharma, Stefan Schaal, Chelsea Finn, Abhishek Gupta, Sergey Levine
An open-source framework that helps facilitate wider usage of RL algorithms and ease deployment to real-world robots.
Daniel Yang, Davin Tjia, Jacob Berg, Dima Damen, Pulkit Agrawal, Abhishek Gupta
Effective shaped rewards can be created from “progress” toward task completion — works in sim, real, and on web-scale video.
I've been a TA at UW for CSE 311 8 times, as well as CSE 142 and CSE 579. In addition, I've been the instructor of record for the following courses:
Introducing students to the shared language and problem-solving toolkit of computer science — the foundation for tackling difficult, real-world computing problems.