Optimal Attacks on Reinforcement Learning Policies.
Alessio RussoAlexandre ProutièrePublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- optimal policy
- cooperative multi agent systems
- optimal control
- control policy
- dynamic programming
- control policies
- total reward
- optimal solution
- state space
- function approximation
- markov decision process
- semi markov decision process
- finite horizon
- expected cost
- markov decision processes
- policy search
- hierarchical reinforcement learning
- reinforcement learning agents
- multi agent
- worst case
- machine learning