Reinforcement Planning for Effective ε-Optimal Policies in Dense Time with Discontinuities.
Léo HenryBlaise GenestAlexandre DreweryPublished in: FSTTCS (2023)
Keyphrases
- optimal policy
- reinforcement learning
- markov decision processes
- decision problems
- dynamic programming
- finite horizon
- state space
- infinite horizon
- initial state
- planning problems
- state dependent
- finite state
- multistage
- sufficient conditions
- partially observable markov decision processes
- optical flow
- dynamic programming algorithms
- serial inventory systems
- machine learning
- long run
- cost function
- markov decision process
- policy iteration
- multi agent
- average reward