Papers
12
Total Citations
326
H-Index
7
About
Noah Siegel is a leading researcher at the intersection of robotics and artificial intelligence, specializing in deep reinforcement learning (deep RL) for complex motor control. His most impactful work demonstrates that deep RL can synthesize sophisticated, agile movement skills for low-cost bipedal robots, as shown in his highly cited 2024 paper on learning soccer skills for a humanoid robot (147 citations). Siegel’s contributions extend to offline and hierarchical RL, where he developed the "Keep Doing What Worked" behavior modeling priors (55+ citations) to improve data efficiency when learning from fixed datasets—a critical advance for real-world robot control. He also pioneered methods for transferring movement skills from human and animal motion capture data to real legged robots, enabling reusable locomotion skills. His work on regularized hierarchical policies and hindsight off-policy option learning further advances compositional transfer and data-efficient skill acquisition. With over 300 total citations, Siegel’s research bridges the gap between simulation and physical robot deployment, making agile, multi-task robot behavior a practical reality.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5Regularized Hierarchical Policies for Compositional Transfer in Robotics.14 citations · 2019
- 6Data-efficient Hindsight Off-policy Option Learning9 citations · 2020
- 7Compositional Transfer in Hierarchical Reinforcement Learning7 citations · 2020
- 8Data-efficient Hindsight Off-policy Option Learning6 citations · 2021
- 9
- 10