CopyCAT:: Taking Control of Neural Policies with Constant Attacks

Léonard Hussenot
Léonard Hussenot

AAMAS '19: International Conference on Autonomous Agents and Multiagent Systems Auckland New Zealand May, 2020, pp. 548-556, 2020.

Cited by: 0|Views21


We propose a new perspective on adversarial attacks against deep reinforcement learning agents. Our main contribution is CopyCAT, a targeted attack able to consistently lure an agent into following an outsider's policy. It is pre-computed, therefore fast inferred, and could thus be usable in a real-time scenario. We show its effectiveness...More



Get fulltext within 24h
Upload PDF

1.Your uploaded documents will be check within 24h, and coins will be credited to your account.

2.As the current system does not support cash withdrawal, you can add staff WeChat (AMxiaomai) to receive it as a red packet.

3.10 coins will be exchanged for 1 yuan.


Upload a single paper

for 5 coins

Wechat's Red Packet

Upload 50 articles

for 280 coins

Wechat's Red Packet

Upload 200 articles

for 1200 coins

Wechat's Red Packet

Upload 500 articles

for 3000 coins

Wechat's Red Packet

Upload 1000 articles

for 7000 coins

Wechat's Red Packet
Your rating :