claim
active
claim:networks-do-not-naturally-decompose-into-architectural-components-such-as-neurons-heads-or-layersNetworks do not naturally decompose into architectural components such as neurons, heads or layers
Motivates new decomposition.
Source paper
extracted_from(2025) · Lee Sharkey · Bilal I. Chughtai · Joshua Batson · Jack W. Lindsey +25
Neighborhood — ranked by edge-count
Papers (1)
paper
Questions (1)
question
- Central open question A.1.1a.1.
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.
- Core claim of the paper: the right level of description for neural representations is geometric structure mirroring the world.
- Foundational for understanding how physiology becomes meaning; decoupling of material state from information content is prerequisite for emergence of cognitive Self.
- Central claim supporting TAME's framework: evidence of cognitive flexibility across diverse embodied systems.
- Architectural requirement from machine learning.
- Lenc & Vedaldi finding on layer-wise alignment
- Central thesis enabling unification of neural, developmental, ecological, and social networks as instances of collective intelligence.
- Alexander's assertion that understanding natural structures at multiple scales is essential to architecture and design philosophy.
- Evidence that evolved machines share biological property of non-optimal modularity, blurring the distinction