Skip to main content

Branching Time Active Inference: The theory and its generality

Champion, Théophile, Da Costa, Lancelot, Bowman, Howard, Grześ, Marek (2022) Branching Time Active Inference: The theory and its generality. Neural Networks, 151 . pp. 295-316. ISSN 0893-6080. (doi:10.1016/j.neunet.2022.03.036) (KAR id:93990)

PDF Publisher pdf
Language: English


Download (962kB) Preview
[thumbnail of 1-s2.0-S0893608022001149-main.pdf]
Preview
This file may not be suitable for users of assistive technology.
Request an accessible format
PDF Author's Accepted Manuscript
Language: English

Restricted to Repository staff only

Contact us about this Publication
[thumbnail of BTAI_Article.pdf]
Official URL:
https://doi.org/10.1016/j.neunet.2022.03.036

Abstract

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.

Item Type: Article
DOI/Identification number: 10.1016/j.neunet.2022.03.036
Uncontrolled keywords: POMDP; Markov process; active inference; variational inference
Subjects: Q Science > QA Mathematics (inc Computing science) > QA276 Mathematical statistics
Q Science > QA Mathematics (inc Computing science) > QA 76 Software, computer programming, > QA76.87 Neural computers, neural networks
Divisions: Divisions > Division of Computing, Engineering and Mathematical Sciences > School of Computing
Depositing User: Marek Grzes
Date Deposited: 11 Apr 2022 21:01 UTC
Last Modified: 31 May 2022 13:16 UTC
Resource URI: https://kar.kent.ac.uk/id/eprint/93990 (The current URI for this page, for reference purposes)
Bowman, Howard: https://orcid.org/0000-0003-4736-1869
Grześ, Marek: https://orcid.org/0000-0003-4901-1539
  • Depositors only (login required):

Downloads

Downloads per month over past year