Andreea Bobu

Berkeley Artificial Intelligence Research
Electrical Engineering and Computer Science
University of California Berkeley
8043-5 Berkeley Way West, 2121 Berkeley Way, Berkeley, CA 94704

I am a Ph.D. student at UC Berkeley, where I work with Anca Dragan in the InterACT Lab. My interests lie at the intersection of machine learning, robotics, and human-robot interaction, with a focus in robot learning with uncertainty. In my research, I tackle ways in which autonomous systems’ models of the world and of other agents (e.g. humans) can go wrong, and improve them for enhanced interaction between people and robots.

I previously received a B.S. in Computer Science and Engineering at MIT in 2017, where I was fortunate to work with Professors Polina Golland and Stefanie Jegelka, and Dr. Adrian Dalca on probabilistic models for medical image analysis.

news

Jun 14, 2022

New paper accepted to Robotics and Automation Letters, "Learning Perceptual Concepts by Bootstrapping from Human Queries". In this work, we ask for human input to learn a low-dimensional variant of the perceptual concept, then use it to generate a larger data set for learning the concept in the high-dimensional space.

May 25, 2022

I gave a talk at Apple's AI/ML seminar.

May 16, 2022
May 13, 2022
Apr 30, 2022

Our paper  "Learning Perceptual Concepts by Bootstrapping from Human Queries" has been accepted as an oral presentation at the  Scaling Robot Learning workshop at ICRA 2022.

Mar 8, 2022

New paper on arXiv, "Teaching Robots to Span the Space of Functional Expressive Motion", where we enable people to efficiently teach robots expressive motions during task execution.

Dec 12, 2021
Nov 16, 2021

I gave a talk at MIT's Interactive Robotics Group.

Nov 3, 2021

I was invited to speak at University of Chicago's Topics in Human-Robot Interaction graduate course.

Jul 23, 2021