Unverified50% confidenceFactExact time
Stable rank is the ratio of the squared Frobenius norm to the squared spectral norm of a matrix, i.e., ||W||_F^2 / ||W||_2^2; it does not exceed the nominal rank of the matrix and is sensitive to singular value decay.
1
Sources
50%
Confidence
Long-term
Relevance
11/2/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe Peter–Weyl theorem decomposes the space L^2(G) of square-integrable functions on a compact group into an orthogonal direct sum of subspaces spanned by the matrix coefficients of all irreducible unitary representations, with each irreducible representation appearing with a multiplicity equal to its dimension.69% similarUnverifiedA Goldstein stationary point is defined via the norm of the gradient mapping: if the supremum of \|\nabla f(x)\| = \|x - prox_{\eta f}(x)\|/\eta does not exceed \delta, then x is called a \delta-Goldstein stationary point.69% similarUnverifiedThe Crouzeix conjecture states: for any complex matrix A and any complex analytic function f on its numerical range, ‖f(A)‖ ≤ 2·sup{|f(z)| : z ∈ W(A)}, with the conjecture that the constant 2 is sharp.68% similarUnverifiedThe derivative of the squaring operation |x|² is 2x, which is linear, so gradients will not explode or vanish sharply at extreme values as they do with exponential functions, giving it a natural advantage in numerical stability.67% similarUnverifiedThe Toeplitz–Hausdorff theorem guarantees that the numerical range of a matrix is always a convex compact set in the complex plane.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1125223API
curl https://kongchang.com/api/v1/knowledge/claims/1125223MCP
get_claim(id=1125223)