framework
active
framework:local-learning-coefficient-llcLocal learning coefficient (LLC)
Singular learning theory measure of loss landscape degeneracy.
Neighborhood — ranked by edge-count
Papers (1)
paper
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.
- Learning rule where change in a parameter at point x,t depends only on system state at same or nearby spacetime points, without requiring global cost function computation
- Hyperparameter for optimizing model parameters through learning in active inference.
- Clarifies what unsupervised learning does.
- Inference of parameters encoding contingencies of the world (e.g., likelihood matrix A) at slower timescale than perception.
- The principle that decision-making about centers must be decentralized to the people closest to them to achieve a living environment.
- How well an RL agent learns, measured by reward curves and associated with causal emergence levels.
- Loss gains a smaller model recovers with infinite data.
- Measures how well an intervened point can be expressed as a convex combination of nearby natural manifold points