How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Self-concordant log-homogeneous barriers and their local geometry
Definition
Let be a finite-dimensional real inner-product space, let be open and convex, and let be . Write for its Hessian as a self-adjoint operator. A self-concordant barrier on is a strictly convex with positive-definite Hessian, with whenever approaches a finite boundary point of , and satisfying both conditions below.
For , define the local norm and its dual by The Dikin inclusion is . The self-concordance inequality is The Newton decrement is .
If is a closed convex cone with nonempty interior and , the barrier is -log-homogeneous when , for every and , and for every . The number is its barrier parameter. When equality constraints are imposed, local norms and decrements are computed after restricting and to the tangent space .
Used by
- Short-step self-concordant path following with finite failure statuses Algorithm
- Log-barrier central paths and primal-dual Newton residuals Definition
- Local Hessian comparison, damped Newton decrement, and center distance Lemma
- The Lorentz logarithmic barrier has parameter two Lemma
- The negative gradient of a log-homogeneous cone barrier is dual feasible Lemma
- The orthant logarithmic barrier has parameter equal to its dimension Lemma
- The positive semidefinite cone is self-dual and its log determinant barrier has parameter n Lemma
- Short-step path following takes O(√νlog(ν/(ε t₀))) Newton steps Theorem
Dependency tree · 0 levels
Nothing. This result depends on no other item in the library.
Sources
- Stephen Boyd and Lieven Vandenberghe, Convex Optimization (standard reference, not scraped)
- Hamza Fawzi, Topics in Convex Optimisation, Lecture 14: Self-concordant functions (standard reference, not scraped)