Saved in:
Bibliographic Details
Main Authors: de Vries, Joery A., He, Jinke, Oren, Yaniv, van der Vaart, Pascal R., de Weerdt, Mathijs M., Spaan, Matthijs T. J.
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.18857
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912917528510464
author de Vries, Joery A.
He, Jinke
Oren, Yaniv
van der Vaart, Pascal R.
de Weerdt, Mathijs M.
Spaan, Matthijs T. J.
author_facet de Vries, Joery A.
He, Jinke
Oren, Yaniv
van der Vaart, Pascal R.
de Weerdt, Mathijs M.
Spaan, Matthijs T. J.
contents Optimally trading-off exploration and exploitation is the holy grail of reinforcement learning as it promises maximal data-efficiency for solving any task. Bayes-optimal agents achieve this, but obtaining the belief-state and performing planning are both typically intractable. Although deep learning methods can greatly help in scaling this computation, existing methods are still costly to train. To accelerate this, this paper proposes a variational framework for learning and planning in Bayes-adaptive Markov decision processes that coalesces variational belief learning, sequential Monte-Carlo planning, and meta-reinforcement learning. In a single-GPU setup, our new method VariBASeD exhibits favorable scaling to larger planning budgets, improving sample- and runtime-efficiency over prior methods.
format Preprint
id arxiv_https___arxiv_org_abs_2602_18857
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle VariBASed: Variational Bayes-Adaptive Sequential Monte-Carlo Planning for Deep Reinforcement Learning
de Vries, Joery A.
He, Jinke
Oren, Yaniv
van der Vaart, Pascal R.
de Weerdt, Mathijs M.
Spaan, Matthijs T. J.
Machine Learning
Optimally trading-off exploration and exploitation is the holy grail of reinforcement learning as it promises maximal data-efficiency for solving any task. Bayes-optimal agents achieve this, but obtaining the belief-state and performing planning are both typically intractable. Although deep learning methods can greatly help in scaling this computation, existing methods are still costly to train. To accelerate this, this paper proposes a variational framework for learning and planning in Bayes-adaptive Markov decision processes that coalesces variational belief learning, sequential Monte-Carlo planning, and meta-reinforcement learning. In a single-GPU setup, our new method VariBASeD exhibits favorable scaling to larger planning budgets, improving sample- and runtime-efficiency over prior methods.
title VariBASed: Variational Bayes-Adaptive Sequential Monte-Carlo Planning for Deep Reinforcement Learning
topic Machine Learning
url https://arxiv.org/abs/2602.18857