Sampling Networks and Aggregate Simulation for Online POMDP Planning.
Hao CuiRoni KhardonPublished in: NeurIPS (2019)
Keyphrases
- planning problems
- partially observable markov decision processes
- online learning
- partially observable markov decision process
- belief space
- partially observable
- heuristic search
- markov decision processes
- planning under uncertainty
- belief state
- finite state
- motion planning
- simulation model
- domain independent
- state space
- social networks
- complex networks
- monte carlo
- spiking neural networks
- decision theoretic
- optimal policy
- low discrepancy sequences
- partially observable stochastic domains
- path finding
- decision problems
- complex systems
- sufficient conditions
- multi agent
- reinforcement learning