Policy Selection

Choosing sequences of actions based on expected free energy; prior probability of policy is softmax of expected free energy

Neighborhood — ranked by edge-count

concept

Policy
related_to
Sequence of actions considered by the agent; basis for planning.
Bayesian Model Averaging
implements
Predictions formed by averaging over policy-specific beliefs, weighted by policy probabilities.

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.

Softmax policy selectionmethod0.801
Selecting policies using a softmax (normalized exponential) function of negative expected free energy.
model selectionconcept0.779
Comparing models using log-evidence approximated by free energy.
Action Selectionconcept0.778
Choice of policies minimizing expected free energy to realize preferred future states.
Selectivitymethod0.764
Adapted control task metric measuring difference between odds-ratio on original task and arbitrary-label control task
Proximal Policy Optimizationmethod0.761
RL algorithm used for training models to comply with the conflicting objective
Sequential Policiesconcept0.748
In active inference, a policy is a sequence of actions through time, as opposed to state-action mappings in RL.
Bayesian Model Selectionconcept0.735
Choosing among candidate models based on model evidence.
How do we pick our goals?question0.725
Points to the meta-cognitive challenge of choosing goal expansion vs. dissolution.