A unified zeroth-order proximal Newton framework for composite optimization establishes iteration and oracle complexity bounds for epsilon-optimality in nonconvex and strongly convex cases, proves local R-superlinear convergence, and shows BFGS is more compatible with finite-difference estimators.
Adaptive first-and zeroth-order methods for weakly convex stochastic optimization problems.arXiv preprint arXiv:2005.09261, 2020
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
Introduces a novel search direction enabling sublinear stochastic bilevel regret guarantees for first- and zeroth-order online bilevel optimization algorithms without relying on window smoothing.
citing papers explorer
-
A Unified Zeroth-Order Proximal Newton-Type Framework for Composite Optimization
A unified zeroth-order proximal Newton framework for composite optimization establishes iteration and oracle complexity bounds for epsilon-optimality in nonconvex and strongly convex cases, proves local R-superlinear convergence, and shows BFGS is more compatible with finite-difference estimators.
-
Stochastic Regret Guarantees for Online Zeroth- and First-Order Bilevel Optimization
Introduces a novel search direction enabling sublinear stochastic bilevel regret guarantees for first- and zeroth-order online bilevel optimization algorithms without relying on window smoothing.