Unverified50% confidenceFactExact time
The researchers derive a closed-form expression for the output-space gap in linear regression and extend it to nonlinear models via the generalized Gauss-Newton method and local linear approximation.
1
Sources
50%
Confidence
Long-term
Relevance
10/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe study explicitly distinguishes between activation geometry and output geometry, emphasizing that it is output geometry, not activation geometry, that is associated with model behavior and controllability.69% similarUnverifiedThe Linear Representation Hypothesis (LRH) assumes that models encode high-level concepts as linear directions in the activation space.66% similarUnverified目前双重下降的严格理论证明主要在线性模型和随机特征模型等简化设定下给出，深度非线性网络的完整数学刻画仍是开放问题64% similarUnverifiedOperator surrogate models are a class of finitely parameterized mappings used to approximate complex (usually nonlinear) operators between infinite-dimensional function spaces.63% similarUnverifiedThis latest theoretical work expands the network output as a generalized Fourier series (GFS) under an orthogonal polynomial basis and writes the evolution of the coefficients under gradient descent as a system of ordinary differential equations, constructing an analytically tractable dynamics matrix H.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1113842API
curl https://kongchang.com/api/v1/knowledge/claims/1113842MCP
get_claim(id=1113842)