Online Convex Optimization in Adversarial Markov Decision Processes.
Aviv RosenbergYishay MansourPublished in: ICML (2019)
Keyphrases
- markov decision processes
- online convex optimization
- long run
- online learning
- optimal policy
- finite state
- reinforcement learning
- convex optimization
- state space
- newton method
- transition matrices
- policy iteration
- efficient algorithms for solving
- dynamic programming
- infinite horizon
- reinforcement learning algorithms
- multi agent
- markov decision process
- average cost
- average reward
- decision theoretic planning
- regret bounds
- lower bound
- support vector