activity
20192026
collaborators
Showing math.OCShow all

6 papers · 1 filter

math.OC2026

Computing Kurdyka-Łojasiewicz exponents via composition and symmetry

Cédric Josz, Wenqing Ouyang

We devise calculus rules for the Kurdyka-Łojasiewicz exponent using the rank theorem and Lie group actions. They apply to a wide class of composite and invariant functions, and are…

math.OC2026

A linesearch-type normal map-based semismooth Newton method for nonsmooth nonconvex composite optimization

Hanfeng Zeng, Wenqing Ouyang, Andre Milzarek

We propose a novel linesearch variant of the trust region normal map-based semismooth Newton method developed in [Ouyang and Milzarek, Math. Program. 212(1-2), 389--435 (2025)] for…

math.OC2026

A MINRES-based Linesearch Algorithm for Nonconvex Optimization with Non-positive Curvature Detection

Hanfeng Zeng, Yang Liu, Wenqing Ouyang +1

We propose a MINRES-based Newton-type algorithm for solving unconstrained nonconvex optimization problems. Our approach uses the minimal residual method (MINRES), a well-known solv…

math.OC2025

Reachability of gradient descent

Cédric Josz, Wenqing Ouyang

We show that gradient descent can converge to any local minimum of a smooth semi-algebraic function. This holds if the step sizes are nonsummable and sufficiently small. The same r…

math.OC2025

Kurdyka-Łojasiewicz exponent via square transformation

Wenqing Ouyang

We consider one of the most common reparameterization techniques, the square transformation. Assuming the original objective function is the sum of a smooth function and a polyhedr…

math.OC2020

Anderson Acceleration for Nonconvex ADMM Based on Douglas-Rachford Splitting

Wenqing Ouyang, Yue Peng, Yuxin Yao +2

The alternating direction multiplier method (ADMM) is widely used in computer graphics for solving optimization problems that can be nonsmooth and nonconvex. It converges quickly t…