optimization pressure

The force of gradient-based learning on structured data that drives networks to organize their representations into geometric structures.

Neighborhood — ranked by edge-count

claim

Geometry arises from optimization pressure on networks trained on structured data.
cites
Mechanistic explanation: geometric structure emerges naturally from standard training on data with underlying structure.

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.

fitness pressureconcept0.822
Evolutionary force favouring solutions like stigmergy to overcome topological limits
Compression pressureconcept0.777
Tanha reframed as the brain's compression drive pushing complexity toward simpler configurations.
Multi-objective optimizationconcept0.761
Framework for optimizing multiple objectives simultaneously, used in MTL.
Deep Optimizationframework0.747
Deontological optimizationconcept0.745
Predictive accuracy applies pressure directly on actions rather than consequences, avoiding instrumental convergence.
In-Context Learning as Optimizationconcept0.734
Direct Preference Optimizationframework0.733
Post-training alignment method during which undesirable behaviors emerged in the studied model.
Parsimony Pressureconcept0.727