Voronoi Progressive Widening: Efficient Online Solvers for Continuous State, Action, and Observation POMDPs
This work provides more efficient and theoretically sound online solvers for continuous POMDPs, which is a significant problem for researchers and practitioners working with complex, real-world decision-making under uncertainty.
This paper addresses the challenge of online solving for Partially Observable Markov Decision Processes (POMDPs) with continuous state, action, and observation spaces. It introduces Voronoi Progressive Widening (VPW) and two algorithms based on it: VOWSS, which provides the first global convergence guarantees for such POMDPs, and VOMCPOW, which consistently outperforms state-of-the-art algorithms in simulations.
This paper introduces Voronoi Progressive Widening (VPW), a generalization of Voronoi optimistic optimization (VOO) and action progressive widening to partially observable Markov decision processes (POMDPs). Tree search algorithms can use VPW to effectively handle continuous or hybrid action spaces by efficiently balancing local and global action searching. This paper proposes two VPW-based algorithms and analyzes them from theoretical and simulation perspectives. Voronoi Optimistic Weighted Sparse Sampling (VOWSS) is a theoretical tool that justifies VPW-based online solvers, and it is the first algorithm with global convergence guarantees for continuous state, action, and observation POMDPs. Voronoi Optimistic Monte Carlo Planning with Observation Weighting (VOMCPOW) is a versatile and efficient algorithm that consistently outperforms state-of-the-art POMDP algorithms in several simulation experiments.