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

CANCloak: Deceiving Two ECUs with One Frame

Li Yue, Zheming Li, Tingting Yin, and Chao Zhang (Tsinghua University)

Read More

Ovid: Message-based Automatic Contact Tracing

Leonie Reichert and Samuel Brack (Humboldt University of Berlin); Björn Scheuermann (Humboldt-University of Berlin)

Read More

CROW: Code Diversification for WebAssembly

Javier Cabrera Arteaga, Orestis Floros, Benoit Baudry, Martin Monperrus (KTH Royal Institute of Technology), Oscar Vera Perez (Univ Rennes, Inria, CNRS, IRISA)

Read More

Bitcontracts: Supporting Smart Contracts in Legacy Blockchains

Karl Wüst (ETH Zurich), Loris Diana (ETH Zurich), Kari Kostiainen (ETH Zurich), Ghassan Karame (NEC Labs), Sinisa Matetic (ETH Zurich), Srdjan Capkun (ETH Zurich)

Read More