On Incorporating Forecasts into Linear State Space Model Markov Decision Processes.
Jacques de ChalendarPeter W. GlynnPublished in: CoRR (2021)
Keyphrases
- pairwise
- markov decision processes
- state space model
- state space
- finite state
- reinforcement learning
- optimal policy
- state estimation
- dynamic programming
- transition matrices
- autoregressive
- action space
- policy iteration
- kalman filter
- infinite horizon
- average cost
- action sets
- markov decision process
- decision theoretic planning
- machine learning
- average reward
- real time dynamic programming
- recursive least squares
- reward function
- kalman filtering
- probability distribution