A Variational Perturbative Approach to Planning in Graph-Based Markov Decision Processes.
Dominik LinznerHeinz KoepplPublished in: AAAI (2020)
Keyphrases
- markov decision processes
- macro actions
- decision theoretic planning
- planning under uncertainty
- partially observable
- state space
- finite state
- optimal policy
- transition matrices
- dynamic programming
- probabilistic planning
- reinforcement learning
- policy iteration
- factored mdps
- risk sensitive
- reachability analysis
- planning problems
- finite horizon
- ai planning
- model based reinforcement learning
- infinite horizon
- average cost
- partially observable markov decision processes
- decision processes
- reward function
- blocks world
- semi markov decision processes
- heuristic search
- action space
- markov decision process
- initial state
- markov decision problems
- average reward
- state abstraction
- sufficient conditions
- optical flow
- reinforcement learning algorithms
- machine learning