Domain Randomization
Domain Randomization is a simulation-based training methodology that systematically varies environmental parameters across a broad distribution during locomotion learning to enhance sim-to-real transfer by increasing policy robustness to domain shifts.
171
Publications
11.6K
Citations
589
Authors
181
Institutions
Publications per year
2017–2026
170
171
Domain randomization for transferring deep neural networks from simulation to the real world
Josh Tobin, Rachel Fong, Alex Ray et al. · 2017 · 2.8K citations
Solving Rubik's Cube with a Robot Hand
Ilge Akkaya, Marcin Andrychowicz, Maciek Chociej et al. · arXiv (Cornell University) · 2019 · 633 citations · Full text
Learning ambidextrous robot grasping policies
Jeffrey Mahler, Matthew Matl, Vishal Satish et al. · Science Robotics · 2019 · 578 citations
1–5 of 171
589
13
h-index
6
h-index
5
h-index
4
h-index
4
h-index
1–5 of 589
181
University of California, Berkeley
Berkeley, United States
15 h-index
80 publications
14.4K citations
Carnegie Mellon University
Pittsburgh, United States
7 h-index
17 publications
974 citations
Google (United States)
Mountain View, United States
7 h-index
18 publications
871 citations
ETH Zurich
Zurich, Switzerland
6 h-index
9 publications
963 citations
Berkeley College
New York, United States
5 h-index
16 publications
5.5K citations

University of Toronto
Toronto, Canada
5 h-index
11 publications
419 citations
Nvidia (United Kingdom)
Reading, United Kingdom
5 h-index
27 publications
1.2K citations
Tongji University
Shanghai, China
4 h-index
8 publications
352 citations
1–8 of 181
Repository · 36 publications · 1.9K citations
21
h-index
7 publications · 281 citations
9
h-index
Conference · 4 publications · 129 citations
5
h-index
1 publications · 27 citations
2
h-index
2 publications · 338 citations
2
h-index
1–5