Saved in:
Bibliographic Details
Main Authors: Bigeard, Antoine, Corso, Anthony, Kochenderfer, Mykel
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.11698
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913844304019456
author Bigeard, Antoine
Corso, Anthony
Kochenderfer, Mykel
author_facet Bigeard, Antoine
Corso, Anthony
Kochenderfer, Mykel
contents Partially observable Markov decision processes (POMDPs) are used to model a wide range of applications, including robotics, autonomous vehicles, and subsurface problems. However, accurately representing the belief is difficult for POMDPs with high-dimensional states. In this paper, we propose a novel approach that uses conditional deep generative models (cDGMs) to represent the belief. Unlike traditional belief representations, cDGMs are well-suited for high-dimensional states and large numbers of observations, and they can generate an arbitrary number of samples from the posterior belief. We train the cDGMs on data produced by random rollout trajectories and show their effectiveness in solving a mineral exploration POMDP with a large and continuous state space. The cDGMs outperform particle filter baselines in both task-agnostic measures of belief accuracy as well as in planning performance.
format Preprint
id arxiv_https___arxiv_org_abs_2505_11698
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Conditional Deep Generative Models for Belief State Planning
Bigeard, Antoine
Corso, Anthony
Kochenderfer, Mykel
Artificial Intelligence
Partially observable Markov decision processes (POMDPs) are used to model a wide range of applications, including robotics, autonomous vehicles, and subsurface problems. However, accurately representing the belief is difficult for POMDPs with high-dimensional states. In this paper, we propose a novel approach that uses conditional deep generative models (cDGMs) to represent the belief. Unlike traditional belief representations, cDGMs are well-suited for high-dimensional states and large numbers of observations, and they can generate an arbitrary number of samples from the posterior belief. We train the cDGMs on data produced by random rollout trajectories and show their effectiveness in solving a mineral exploration POMDP with a large and continuous state space. The cDGMs outperform particle filter baselines in both task-agnostic measures of belief accuracy as well as in planning performance.
title Conditional Deep Generative Models for Belief State Planning
topic Artificial Intelligence
url https://arxiv.org/abs/2505.11698