framework
active
framework:brill-capacity-allocation-modelBrill capacity-allocation model
Models allocate capacity among power-law-distributed data manifolds.
Neighborhood — ranked by edge-count
Papers (1)
paper
Thinkers (1)
thinker
- Ari BrillstudiesAuthor of the neural scaling model adapted here.
Frameworks (1)
framework
- SAE latent allocation modelextendsReduces SAE optimization to allocating N latents across features.
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.
- Analytical result showing exponential power activation allows memory storage scaling as 2^(N/2); cited in context of Hopfield scaling.
- Bigger models are more likely to converge to a shared representation than smaller models because they can better approximate the global optimum
- Links toy model to practice.
- Open-ended capacity for associative learning proposed as a transition marker for consciousness.
- Proposed mechanism for exclusive-feature density.
- Caveat and forward-looking statement from the abstract.
- Mean-field theory model describing phase transitions in wealth distribution with economic growth.
- Once recognized, the self/environment partition appears not as a given fact but as an optional modelling decision.