Papers
12
Total Citations
210
H-Index
8
About
Stefan Bauer is a researcher at the forefront of robotics, reinforcement learning, and representation learning, whose work has significantly advanced the field of dexterous robotic manipulation and transfer learning. Best known for developing the **TriFinger** robotic platform—an open-source system designed to democratize dexterity research by reducing the prohibitive costs of real-world experimentation—Bauer has helped establish a rigorous experimental foundation for the broader robotics community. His pioneering work on sim-to-real transfer, particularly in leveraging GPU-accelerated simulation environments like NVIDIA's IsaacGym to train policies deployable on physical robots, has garnered substantial attention, with his 2022 manipulation paper alone accumulating 45 citations. Beyond hardware, Bauer has made meaningful contributions to causal and structured learning through **CausalWorld**, a benchmark environment enabling researchers to study causal structure and skill transfer in robotic manipulation settings. His work on disentanglement datasets further bridges the gap between synthetic training environments and real-world applicability. Collectively spanning over 200 citations, Bauer's research reflects a coherent vision: building robotic agents that learn efficiently, generalize robustly, and ultimately operate reliably outside controlled laboratory conditions—a challenge central to the future of intelligent autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5TriFinger: An Open-Source Robot for Learning Dexterity24 citations · 2020
- 6
- 7
- 8
- 9
- 10