Lukas Krauch
Papers
1
Total Citations
5
H-Index
1
About
Lukas Krauch is a researcher at the forefront of sample-efficient reinforcement learning (RL) for real-world robotics, with a particular focus on high-speed, dynamic manipulation tasks. His most-cited work, "Sample-efficient Reinforcement Learning in Robotic Table Tennis" (2021, 5 citations), addresses a critical bottleneck in applying RL to physical systems: the need for vast numbers of training episodes. By developing algorithms that learn effectively from far fewer attempts, Krauch’s research bridges the gap between simulation-based RL successes and practical robotic applications where each trial is costly and time-consuming. His contributions are especially notable for enabling robots to master complex, reactive skills—like returning a ping-pong ball—with minimal real-world interaction. This work not only advances the field of robot learning but also provides a blueprint for deploying RL in settings where data efficiency is paramount, such as manufacturing, surgery, and autonomous systems. Krauch’s research is a key step toward making RL a viable tool for real-world, high-performance robotics.
Research Focus
Key Achievements
Top Papers
- 1Sample-efficient Reinforcement Learning in Robotic Table Tennis5 citations · 2021