concept
active
concept:model-selectionmodel selection
Comparing models using log-evidence approximated by free energy.
Neighborhood — ranked by edge-count
Concepts (1)
concept
- Bayesian Model Selectionrelated_toChoosing among candidate models based on model evidence.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- A representation that captures relevant aspects of a system; according to the theorem, the regulator must embody this.
- A model trained on comparison data to assign scores to responses, used as reward signal in RLHF/RLAIF.
- Analogy between evolution (model selection) and Bayesian model reduction; but evolution is not curious or insightful
- Used as theoretical motivation for UCCT's log k budget term as complexity penalty mirroring model selection
- Using interventions to guide model generation behavior, e.g., adding sentiment vectors at inference time
- Adapted control task metric measuring difference between odds-ratio on original task and arbitrary-label control task
- Technique for modifying model knowledge or behavior via targeted interventions, e.g., ROME by Meng et al.
- Choice of policies minimizing expected free energy to realize preferred future states.