Training Adversarial Agents to Exploit Weaknesses in Deep Control Policies.

Sampo Kuutti Saber Fallah Richard Bowden

Published in: ICRA (2020)

Keyphrases

control policies
multi agent
multi agent systems
multiple agents
cooperative
reinforcement learning
stochastic optimization problems
motion control
optimal policy
control system
control policy
action space
decision making
real world
finite horizon
real time
search algorithm