Towards Causal Temporal Reasoning for Markov Decision Processes.
Milad KazemiNicola PaolettiPublished in: CoRR (2022)
Keyphrases
- temporal reasoning
- markov decision processes
- temporal constraints
- temporal information
- qualitative and quantitative
- optimal policy
- state space
- reinforcement learning
- transition matrices
- finite state
- policy iteration
- dynamic programming
- temporal knowledge
- event calculus
- temporal constraint networks
- average reward
- constraint networks
- partially observable
- bayesian networks
- average cost
- markov decision process
- planning under uncertainty
- reachability analysis
- action space
- decision theoretic planning
- reward function
- infinite horizon
- computational complexity
- action sets
- model based reinforcement learning
- sufficient conditions