Bayesian Model Expansion

Adding new states or parameters to the generative model if it increases model evidence, enabling concept learning.

Neighborhood — ranked by edge-count

concept

Structure Learning
implements
Updating the structure of the generative model to better account for observations via Bayesian model reduction and expansion.

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

Bayesian model expansion allows for generalisation and concept learning in active inference.claim0.853
Definition of Bayesian model expansion, Section 9.2.
Bayesian Model Selectionconcept0.842
Choosing among candidate models based on model evidence.
Bayesian model evidenceconcept0.818
The probability of sensory data under a generative model; negative log evidence is bounded by free energy.
Bayesian Model Averagingconcept0.810
Predictions formed by averaging over policy-specific beliefs, weighted by policy probabilities.
Bayesian Model of Painframework0.786
Conceptualization of pain perception as inference over hidden nociceptive causes, from Eckert et al. 2022
Bayesian model reduction (method)method0.785
A method for simplifying models by removing parameters that don't contribute; applied to eliminate the self-boundary prior.
Natural Selection as Bayesian Model Selectionconcept0.781
Analogy between evolution (model selection) and Bayesian model reduction; but evolution is not curious or insightful
Bayesian inversionconcept0.778
Online inversion of nonlinear dynamic causal models using DEM.