Zhikun Wang

Zhikun Wang joined the Max-Planck Institute of Biological Cybernetics in 2009 as a Ph.D. student in the Department of Bernhard Schölkopf under the supervision of Jan Peters and Christoph Lampert. Before doing so, he studied at the Tsinghua University in China, where he received his Bachelor’s and Master’s degrees in Computer Science.

Zhikun Wang is currently working on the learning to play ping-pong project. He uses a set of cameras to capture the opponent’s movement, track the paddle trajectory and predict how the ball is likely to be returned. He also works on two-agent reinforcement learning algorithms. He tries to model the opponents’ policy and learn a robust strategy. In the future, this work will be applied to learn a ping-pong playing strategy for the robot.

Key References

  1. Wang, Z.; Boularias, A.; Muelling, K.; Peters, J. (2011). Balancing Safety and Exploitability in Opponent Modeling, Proceedings of the Twenty-Fourth National Conference on Artificial Intelligence (AAAI)  download [PDF]
  2. Wang, Z.; Lampert, C; Muelling, K; Schoelkopf, B.; Peters, J. (2011). Learning Anticipation Policies for Robot Table Tennis, IEEE/RSJ International Conference on Intelligent Robot Systems (IROS)  download [PDF]

 

 

Contact Information

Mail: Zhikun Wang, Spemannstr. 38, 72076 Tübingen, Germany
work +49-7071-601-556

zum Seitenanfang