I’m a second-year PhD student in computer science at UC Berkeley advised by Ben Recht.
Previously, I received my Bachelor’s in electrical engineering and computer science from MIT and my Master’s in human rights studies from Columbia University.
I work on on interdisciplinary applications of computer science, from
astrophysics to history to politics.
My research has been featured in over 100 media outlets and has been
liked and shared tens of thousands of times on social of media.
Email
Curriculum Vitae
Google Scholar
01 EVOLUTION GYM2021
ABSTRACT:
Both the design and control of a robot play equally important roles
in its task performance. However, while optimal control is well studied
in the machine learning and robotics community, less attention is
placed on finding the optimal robot design. This is mainly because
co-optimizing design and control in robotics is characterized as a
challenging problem, and more importantly, a comprehensive evaluation
benchmark for co-optimization does not exist.
In this paper, we propose
Evolution Gym, the first large-scale benchmark for co-optimizing the
design and control of soft robots. In our benchmark, each robot is
composed of different types of voxels (e.g., soft, rigid, actuators),
resulting in a modular and expressive robot design space. Our benchmark
environments span a wide range of tasks, including locomotion on various
types of terrains and manipulation. Furthermore, we develop several
robot co-evolution algorithms by combining state-of-the-art design
optimization methods and deep reinforcement learning techniques.
Evaluating the algorithms on our benchmark platform, we observe robots
exhibiting increasingly complex behaviors as evolution progresses, with
the best evolved designs solving many of our proposed tasks.
Additionally, even though robot designs are evolved autonomously from
scratch without prior knowledge, they often grow to resemble existing
natural creatures while outperforming hand-designed robots.
Nevertheless, all tested algorithms fail to find robots that succeed in
our hardest environments. This suggests that more advanced algorithms
are required to explore the high-dimensional design space and evolve
increasingly intelligent robots – an area of research in which we hope
Evolution Gym will accelerate progress.
Our website with code,
environments, documentation, and tutorials is available here.
PUBLICATION:
Evolution Gym: A Large-Scale Benchmark for Evolving Soft Robots
Jagdeep Singh Bhatia, Holly Jackson, Yunsheng Tian, Jie Xu, Wojciech Matusik
Thirty-fifth Conference on Neural Information Processing Systems (NeurIPS), 2021
Paper
Code
Docs
Poster
Video
SELECTED PRESS:
MIT News
WIRED
IEEE Spectrum