Yunzhe Tian, Yike Li, Yingxiao Xiang, Wenjia Niu, Endong Tong, and Jiqiang Liu (Beijing Jiaotong University)

Robust reinforcement learning has been a challenging problem due to always unknown differences between real and training environment. Existing efforts approached the problem through performing random environmental perturbations in learning process. However, one can not guarantee perturbation is positive. Bad ones might bring failures to reinforcement learning. Therefore, in this paper, we propose to utilize GAN to dynamically generate progressive perturbations at each epoch and realize curricular policy learning. Demo we implemented in unmanned CarRacing game validates the effectiveness.

View More Papers

Why Do Programmers Do What They Do? A Theory...

Lavanya Sajwan, James Noble, Craig Anslow (Victoria University of Wellington), Robert Biddle (Carleton University)

Read More

Scenario-Driven Assessment of Cyber Risk Perception at the Security...

Simon Parkin (TU Delft), Kristen Kuhn, Siraj Ahmed Shaikh (Coventry University)

Read More

Short Paper: Declarative Demand-Driven Reverse Engineering

Yihao Sun, Jeffrey Ching, Kristopher Micinski (Department of Electical Engineering and Computer Science, Syracuse University)

Read More