Active inference and artificial reasoning

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Friston, Karl, Da Costa, Lancelot, Tschantz, Alexander, Heins, Conor, Buckley, Christopher, Verbelen, Tim, Parr, Thomas
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917168932716544
author Friston, Karl
Da Costa, Lancelot
Tschantz, Alexander
Heins, Conor
Buckley, Christopher
Verbelen, Tim
Parr, Thomas
author_facet Friston, Karl
Da Costa, Lancelot
Tschantz, Alexander
Heins, Conor
Buckley, Christopher
Verbelen, Tim
Parr, Thomas
contents This technical note considers the sampling of outcomes that provide the greatest amount of information about the structure of underlying world models. This generalisation furnishes a principled approach to structure learning under a plausible set of generative models or hypotheses. In active inference, policies - i.e., combinations of actions - are selected based on their expected free energy, which comprises expected information gain and value. Information gain corresponds to the KL divergence between predictive posteriors with, and without, the consequences of action. Posteriors over models can be evaluated quickly and efficiently using Bayesian Model Reduction, based upon accumulated posterior beliefs about model parameters. The ensuing information gain can then be used to select actions that disambiguate among alternative models, in the spirit of optimal experimental design. We illustrate this kind of active selection or reasoning using partially observed discrete models; namely, a 'three-ball' paradigm used previously to describe artificial insight and 'aha moments' via (synthetic) introspection or sleep. We focus on the sample efficiency afforded by seeking outcomes that resolve the greatest uncertainty about the world model, under which outcomes are generated.
format Preprint
id arxiv_https___arxiv_org_abs_2512_21129
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Active inference and artificial reasoning
Friston, Karl
Da Costa, Lancelot
Tschantz, Alexander
Heins, Conor
Buckley, Christopher
Verbelen, Tim
Parr, Thomas
Neurons and Cognition
Data Analysis, Statistics and Probability
Machine Learning
This technical note considers the sampling of outcomes that provide the greatest amount of information about the structure of underlying world models. This generalisation furnishes a principled approach to structure learning under a plausible set of generative models or hypotheses. In active inference, policies - i.e., combinations of actions - are selected based on their expected free energy, which comprises expected information gain and value. Information gain corresponds to the KL divergence between predictive posteriors with, and without, the consequences of action. Posteriors over models can be evaluated quickly and efficiently using Bayesian Model Reduction, based upon accumulated posterior beliefs about model parameters. The ensuing information gain can then be used to select actions that disambiguate among alternative models, in the spirit of optimal experimental design. We illustrate this kind of active selection or reasoning using partially observed discrete models; namely, a 'three-ball' paradigm used previously to describe artificial insight and 'aha moments' via (synthetic) introspection or sleep. We focus on the sample efficiency afforded by seeking outcomes that resolve the greatest uncertainty about the world model, under which outcomes are generated.
title Active inference and artificial reasoning
topic Neurons and Cognition
Data Analysis, Statistics and Probability
Machine Learning
url https://arxiv.org/abs/2512.21129