hypothesis

active

hypothesis:if-epistemic-value-is-removed-from-expected-free-energy-the-resulting-objective-reduces-to-maximizing-expected-future-reward-pragmatic-value

If epistemic value is removed from expected free energy, the resulting objective reduces to maximizing expected future reward (pragmatic value).

Stated as conditional statement explaining the special case whence RL emerges.

Source paper

extracted_from

Active inference: demystified and compared

(2021) · Noor Sajid · Philip J. Ball · Thomas Parr · Karl J. Friston

Neighborhood — ranked by edge-count

Claims (1)

claim

Reinforcement learning can be regarded as a limiting or special case of model-based approaches in general — or active inference in particular — when epistemic value is removed.
supports
§3 Discussion.

Related by similarity (8)

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.

Expected free energy decomposes into risk (exploitation) and ambiguity (exploration) terms, providing optimal balance between goal-seeking and novelty-seeking.claim0.814
Key insight into structure of decision-making; explains intrinsic motivation and curiosity.
Minimising expected free energy specifies the optimal balance between goal-seeking and itinerant novelty-seeking behaviour.claim0.806
Describes the exploitation-exploration trade-off.
Minimising expected free energy ensures preferred outcomes are realised, thereby avoiding surprise on average.claim0.802
How expected free energy links to survival.
Acting to optimize value and perception are two aspects of exactly the same principle: minimization of free energy.claim0.795
Foundational claim unifying action and perception within single optimization framework.
Expected free energy scores each possible course of action in terms of its ability to reach the range of preferred states of being.claim0.793
Definitional claim from Section 2.
Curiosity, insight, decision-making, and diverse phenomena can all be accommodated by a single imperative: minimization of expected free energy (resolution of uncertainty).claim0.790
Central thesis of the paper unifying cognitive phenomena under one objective function
The most likely (i.e., best) policies minimise expected free energy.claim0.785
Decision-making rule in active inference.
The expected free energy can be unpacked in terms of risk and ambiguity.claim0.782
A key decomposition presented in Section 7.