Policy Evaluation in Decentralized POMDPs with Belief Sharing.
Mert KayaalpFatima GhadiehAli H. SayedPublished in: CoRR (2023)
Keyphrases
- policy evaluation
- partially observable markov decision processes
- belief state
- reinforcement learning
- markov decision processes
- point based value iteration
- multi agent
- least squares
- state space
- policy iteration
- finite state
- temporal difference
- belief revision
- optimal policy
- planning under uncertainty
- partially observable
- model free
- monte carlo
- dynamical systems
- dynamic programming
- variance reduction
- policy gradient
- decision problems
- function approximation
- planning problems
- infinite horizon
- belief functions
- dynamic bayesian networks
- markov decision process
- computational complexity
- approximation methods
- average reward
- markov decision problems
- cost function