Ryan Hoque

I am a third year PhD candidate in the UC Berkeley EECS department working on Robotics and Artificial Intelligence. I am advised by Ken Goldberg and am part of the Berkeley Artificial Intelligence Research (BAIR) lab. I have also collaborated with Pieter Abbeel and folks from Google Brain, Meta AI, Honda Research Institute, and Uber ATG (acq. Aurora). My research primarily focuses on algorithms for robot fleet learning from interactive human supervision.

Email  /  CV  /  Google Scholar  /  Github  /  LinkedIn  /  Twitter

My PhD research is focused on the development of interactive imitation and reinforcement learning algorithms that scale to large robot fleets performing complex tasks (e.g., manipulation). In my undergraduate and Master's research, I worked on learning algorithms for robotic manipulation of deformable objects. In addition to selected publications below (see my CV for the full list), check out:

• My talk at CoRL 2022 in Auckland, New Zealand and BAIR Blog post on Interactive Fleet Learning. This is ideal if you're looking for a quick and accessible highlight of my recent work.
• Issue #238 of the Import AI newsletter, read by more than 10,000 AI experts, featuring our work at the top.
• University courses that have featured my research, including CS 16-881: Deep Reinforcement Learning for Robotics at Carnegie Mellon University, CS 6960: Human-AI Alignment at the University of Utah, and CS 343: Artificial Intelligence at the University of Texas, Austin.

We introduce new formalism, algorithms, and open-source benchmarks for "Interactive Fleet Learning": interactive learning with multiple robots and multiple humans.

A novel interactive imitation learning algorithm that reasons about both state novelty and risk to actively query for human interventions more efficiently than prior algorithms.

We propose "context switching" between human and robot control as a measure of supervisor burden in interactive imitation learning, and propose an algorithm that reduces context switching during training and execution.

(In collaboration with Robotics at Google) We perform the first systematic benchmarking of fabric manipulation algorithms with Google Reach, a prototype hardware testbed for low-latency remote robot control over the Internet.

(In collaboration with Honda Research Institute) Journal paper extending the VSF conference paper below with new datasets, visual dynamics models, cost functions, optimizers, and physical experiments.

(In collaboration with Honda Research Institute) A novel model-based reinforcement learning technique that trains a visual dynamics model for sequentially manipulating fabric toward a variety of goal images, entirely from random interaction RGBD data in simulation.

I'm a Bay Area native and a lifelong bear: I earned my B.S. and M.S. in EECS at UC Berkeley and am still here for my PhD. Outside of research, I enjoy an eclectic mix of hobbies including playing the piano and ukulele, reading and writing about philosophy (especially metaphysics), and the outdoors. Some fun facts you may find interesting:

Website template from BAIR alum Jon Barron. This template seems to be quite popular!