Bayesian Model Averaging

Predictions formed by averaging over policy-specific beliefs, weighted by policy probabilities.

Neighborhood — ranked by edge-count

thinker

concept

Prediction Error
supports
Role in optimizing sensory states; unified treatment shows value-learning and perception share error-minimization principle.
Policy Selection
implements
Choosing sequences of actions based on expected free energy; prior probability of policy is softmax of expected free energy

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 evidenceconcept0.814
The probability of sensory data under a generative model; negative log evidence is bounded by free energy.
Bayesian Model Selectionconcept0.813
Choosing among candidate models based on model evidence.
Bayesian Model Expansionmethod0.810
Adding new states or parameters to the generative model if it increases model evidence, enabling concept learning.
Bayesian Model-Based Reinforcement Learningframework0.791
RL variant that maintains beliefs over environment model; compared to active inference using Thompson sampling.
Bayesian Smoothingmethod0.780
State estimation that combines prior expectations with likelihood; updates informed by past and future states.
Bayesian Model of Painframework0.769
Conceptualization of pain perception as inference over hidden nociceptive causes, from Eckert et al. 2022
Autoregressive modelsframework0.763
Second model system studied; used to show why flat autoregressive LLMs struggle with long-range coherence.
Bayesian model reduction (method)method0.763
A method for simplifying models by removing parameters that don't contribute; applied to eliminate the self-boundary prior.