Certificate-Guided Pruning for Stochastic Lipschitz Optimization

ORID 9CqZoRWpoc · tags icml2026-repro paper-9CqZoRWpoc

#StatusPageArtifactClaim excerpt
1VERIFIED 2/201-certificate-guided-pruning-cgp-maintains-expliciartifactCertificate-Guided Pruning (CGP) maintains an explicit active set A_t of candida…
2VERIFIED 2/202-margin-condition-near-optimality-dimension-assumartifactUnder a margin condition with near-optimality dimension α (Assumption 2.3), the …
3VERIFIED 2/203-cgp-optimality-probability-least-logartifactCGP achieves ε-optimality with probability at least 1−δ using T = Õ(L^d ε^{-(2+α…
4VERIFIED 2/204-matching-lower-bound-any-algorithmartifactA matching lower bound shows any algorithm requires Ω(ε^{-(2+α)}) samples under …
5VERIFIED 2/205-cgp-adaptive-learns-lipschitz-constant-onlineartifactCGP-Adaptive learns the Lipschitz constant L online via a doubling scheme, addin…
6VERIFIED 2/206-cgp-tr-trust-region-variant-scales-dimensionartifactCGP-TR, a trust-region variant, scales to dimension d > 50 via certified restart…

Open logbook index · logbook.json