Improved Planning for Infinite-Horizon Interactive POMDPs using Probabilistic Inference (Extended Abstract).
Xia QuPrashant DoshiPublished in: AAMAS (2015)
Keyphrases
- extended abstract
- probabilistic inference
- infinite horizon
- partially observable
- partially observable markov decision processes
- production planning
- dec pomdps
- markov decision processes
- optimal policy
- finite horizon
- markov decision problems
- graphical models
- optimal control
- long run
- bayesian networks
- dynamic programming
- stochastic demand
- weighted model counting
- reinforcement learning
- message passing
- markov decision process
- approximate inference
- state space
- belief networks
- conditional probabilities
- influence diagrams
- average cost
- planning problems
- planning under uncertainty
- belief state
- dynamical systems
- control system
- linear programming
- lead time
- np hard
- control strategy
- special case
- decision theoretic
- policy iteration
- multi agent
- heuristic search
- machine learning
- decision problems