Training Adversarial Agents to Exploit Weaknesses in Deep Control Policies.
Sampo KuuttiSaber FallahRichard BowdenPublished in: CoRR (2020)
Keyphrases
- control policies
- multi agent
- multi agent systems
- multiple agents
- stochastic optimization problems
- reinforcement learning
- action space
- cooperative
- control policy
- control strategies
- optimal policy
- motion control
- supervised learning
- finite horizon
- markov decision processes
- single agent
- learning algorithm
- state space
- decision making