• Home
  • Search
  • Boosting MCTS With Free Energy Minimization.
  • https://doi.org/10.1162/neco.a.31Copy DOI Icon

Boosting MCTS With Free Energy Minimization.

Show More
  • Abstract
  • Literature Map
  • References
  • Similar Papers
Abstract

Active inference, grounded in the free energy principle, provides a powerful lens for understanding how agents balance exploration and goal-directed behavior in uncertain environments. Here, we propose a new planning framework that integrates Monte Carlo tree search (MCTS) with active inference objectives to systematically reduce epistemic uncertainty while pursuing extrinsic rewards. Our key insight is that MCTS, already renowned for its search efficiency, can be naturally extended to incorporate free energy minimization by blending expected rewards with information gain. Concretely, the cross-entropy method (CEM) is used to optimize action proposals at the root node, while tree expansions leverage reward modeling alongside intrinsic exploration bonuses. This synergy allows our planner to maintain coherent estimates of value and uncertainty throughout planning, without sacrificing computational tractability. Empirically, we benchmark our planner on a diverse set of continuous control tasks, where it demonstrates performance gains over both stand-alone CEM and MCTS with random rollouts.

Similar Papers
  • Conference Article
  • Citations3

Monte Carlo Tree Search with Variable Simulation Periods for Continuously Running Tasks

  • Nov 01, 2019
  • Seydou Ba +4
  • Conference Article

Environmental Map Learning with Multiple-Robots

  • May 19, 2025
  • Azin Shamshirgaran +1
  • Research Article
  • Citations2

Branching Time Active Inference with Bayesian Filtering.

  • Sep 12, 2022
  • Neural Computation
  • Théophile Champion +2
  • Discussion
  • Citations11

Free energy minimization and information gain: The devil is in the details

  • Jul 27, 2015
  • Cognitive Neuroscience
  • Johan Kwisthout +1
  • Research Article
  • Citations25

Feature Selection for High Dimensional Data Using Monte Carlo Tree Search

  • Jan 01, 2018
  • IEEE Access
  • Muhammad Umar Chaudhry +1
  • PDF
  • Research Article
  • Citations28

Deep Active Inference and Scene Construction.

  • Oct 28, 2020
  • Frontiers in artificial intelligence
  • R Conor Heins +5
  • Conference Article
  • Citations1

ConstrainedZero: Chance-Constrained POMDP Planning Using Learned Probabilistic Failure Surrogates and Adaptive Safety Constraints

  • Aug 01, 2024
  • Chao Xu +3
  • PDF
  • Research Article
  • Citations19

Forgetting ourselves in flow: an active inference account of flow states and how we experience ourselves within them.

  • Jun 03, 2024
  • Frontiers in psychology
  • Darius Parvizi-Wayne +5
  • Book Chapter

Free-Energy Principle and Predictive Coding: A Computational Theory Explaining Various Brain Functions

  • Jan 01, 2026
  • Shingo Murata
  • Preprint Article

The Precision-Weighting Dysregulation Hypothesis: An Active Inference Framework for Criminal Behaviour, Differential Pathways, and Testable Predictions

  • Mar 04, 2026
  • Preprints.org
  • Fritz Hemmerich
  • PDF
  • Research Article
  • Citations43

An Introduction to Predictive Processing Models of Perception and Decision-Making.

  • Oct 29, 2023
  • Topics in Cognitive Science
  • Mark Sprevak +1
  • Research Article
  • Citations23

The free energy principle: it’s not about what it takes, it’s about what took you there

  • Feb 22, 2021
  • Biology & Philosophy
  • Axel Constant
  • Research Article
  • Citations10

Branching time active inference: Empirical study and complexity class analysis

  • May 20, 2022
  • Neural Networks
  • Théophile Champion +2
  • Research Article
  • Citations188

Active Inference: Demystified and Compared.

  • Jan 05, 2021
  • Neural Computation
  • Noor Sajid +3
  • PDF
  • Research Article
  • Citations39

Learning Generative State Space Models for Active Inference.

  • Nov 16, 2020
  • Frontiers in Computational Neuroscience
  • Ozan Çatal +4
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.