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

Does Every Second Count? Time-based Evolution of Malware Behavior...

Alexander Küchler (Fraunhofer AISEC), Alessandro Mantovani (EURECOM), Yufei Han (NortonLifeLock Research Group), Leyla Bilge (NortonLifeLock Research Group), Davide Balzarotti (EURECOM)

Read More

Understanding and Detecting International Revenue Share Fraud

Merve Sahin (SAP Security Research), Aurélien Francillon (EURECOM)

Read More

OblivSketch: Oblivious Network Measurement as a Cloud Service

Shangqi Lai (Monash University), Xingliang Yuan (Monash University), Joseph K. Liu (Monash University), Xun Yi (RMIT University), Qi Li (Tsinghua...

Read More

Data Poisoning Attacks to Deep Learning Based Recommender Systems

Hai Huang (Tsinghua University), Jiaming Mu (Tsinghua University), Neil Zhenqiang Gong (Duke University), Qi Li (Tsinghua University), Bin Liu (West...

Read More