mlsanityData to 14 Oct 2026

Branching Time Active Inference: the theory and its generality

arXiv 2111.11107 ↗First posted 22 Nov 2021cs.AIcited by 15

Not on the map. Why?

Abstract (arXiv)

Over the last 10 to 15 years, active inference has helped to explain various brain mechanisms from habit formation to dopaminergic discharge and even modelling curiosity. However, the current implementations suffer from an exponential (space and time) complexity class when computing the prior over all the possible policies up to the time-horizon. Fountas et al (2020) used Monte Carlo tree search to address this problem, leading to impressive results in two different tasks. In this paper, we present an alternative framework that aims to unify tree search and active inference by casting planning as a structure learning problem. Two tree search algorithms are then presented. The first propagates the expected free energy forward in time (i.e., towards the leaves), while the second propagates it backward (i.e., towards the root). Then, we demonstrate that forward and backward propagations are related to active inference and sophisticated inference, respectively, thereby clarifying the differences between those two planning strategies.

Key facts. Cited by 15 papers in the map (data through 14 Oct 2026). Builds on 16 of its 57 references that are in the map. First posted on arXiv 22 Nov 2021.

Builds on

16 of 57 references.

References · 16

  1. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  2. 2013Playing Atari with Deep Reinforcement Learningcited by 3,274
  3. 2015Deep Reinforcement Learning with Double Q-learningcited by 1,745
  4. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,353
  5. 2017A Tutorial on Thompson Samplingcited by 433
  6. 2016Playing FPS Games with Deep Reinforcement Learningcited by 95
  7. 2020Active inference on discrete state-spaces: a synthesiscited by 92
  8. 2020Deep active inference agents using Monte-Carlo methodscited by 72
  9. 2020Sophisticated Inferencecited by 56
  10. 2019Deep Active Inference as Variational Policy Gradientscited by 53
  11. 2020Whence the Expected Free Energy?cited by 40
  12. 2020Action and Perception as Divergence Minimizationcited by 32
4 more
  1. 2019End-to-End Pixel-Based Deep Active Inference for Body Perception and Actioncited by 24
  2. 2020Active Inference and Behavior Trees for Reactive Action Planning and Execution in Roboticscited by 23
  3. 2020Learning Perception and Planning with Deep Active Inferencecited by 19
  4. 2018A Factor Graph Approach to Automated Design of Bayesian Signal Processing Algorithmscited by 11

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 1; 2022: 3; 2023: 3; 2024: 1; 2025: 4; 2026: 3.

15 papers.

Sort:

By area:

  1. 2023On efficient computation in active inferencecited by 15
  2. 2024Active Inference as a Model of Agencycited by 12
  3. 2025Expected Free Energy-based Planning as Variational Inferencecited by 10
  4. 2022Geometric Methods for Sampling, Optimisation, Inference and Adaptive Agentscited by 5
  5. 2023Deconstructing deep active inferencecited by 5
  6. 2025A Message Passing Realization of Expected Free Energy Minimizationcited by 5
  7. 2023Realising Synthetic Active Inference Agents, Part I: Epistemic Objectives and Graphical Specification Languagecited by 4
  8. 2021Branching Time Active Inference with Bayesian Filteringcited by 2
  9. 2022Multi-Modal and Multi-Factor Branching Time Active Inferencecited by 2
  10. 2026What Type of Inference is Active Inference?cited by 2
  11. 2022Efficient search of active inference policy spaces using k-meanscited by 0
  12. 2025Boosting MCTS with Free Energy Minimizationcited by 0
3 more
  1. 2025A memory-based active inference model of DishBrain-like adaptive behaviourcited by 0
  2. 2026Expected Free Energy-based Planning as Variational Inferencecited by 0
  3. 2026Sophisticated Policies from Epistemic Priorscited by 0