On exploration of an interior mirror descent flow for stochastic nonconvex constrained problem
Not on the map. Why?
Abstract (arXiv)
We study a nonsmooth nonconvex optimization problem defined over nonconvex constraints, where the feasible set is given by the intersection of the closure of an open set and a smooth manifold. By endowing the open set with a Riemannian metric induced by a barrier function, we obtain a Riemannian subgradient flow formulated as a differential inclusion, which remains strictly within the interior of the feasible set. This continuous dynamical system unifies two classes of iterative optimization methods, namely the Hessian barrier method and mirror descent scheme, by revealing that these methods can be interpreted as discrete approximations of the continuous flow. We explore the long-term behavior of the trajectories generated by this dynamical system and show that the existing deficient convergence properties of the Hessian barrier and mirror descent scheme can be unifily and more insightfully interpreted through these of the continuous trajectory. For instance, the notorious spurious stationary points citechen2024spurious observed in Hessian barrier method and mirror descent scheme are interpreted as stable equilibria of the dynamical system that do not correspond to real stationary points of the original optimization problem. We provide two sufficient condition such that these spurious stationary points can be avoided if the strict complementarity conditions holds. In the absence of these regularity condition, we propose a random perturbation strategy that ensures the trajectory converges (subsequentially) to an approximate stationary point. Building on these insights, we introduce two iterative Riemannian subgradient methods, form of interior point methods, that generalizes the existing Hessian barrier method and mirror descent scheme for solving nonsmooth nonconvex optimization problems.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 32 of its 69 references that are in the map. First posted on arXiv 21 Jul 2025.
Builds on
32 of 69 references.
References · 32
- 2018Stochastic subgradient method converges on tame functionscited by 124
- 2016Relatively-Smooth Convex Optimization by First-Order Methods, and Applicationscited by 108
- 2019Conservative set valued fields, automatic differentiation, stochastic gradient method and deep learningcited by 72
- 2017Stochastic Methods for Composite and Weakly Convex Optimization Problemscited by 48
- 2022Implicit Bias of Gradient Descent on Reparametrized Models: On Equivalence to Mirror Descentcited by 39
- 2017First Order Methods beyond Convexity and Lipschitz Gradient Continuity with Applications to Quadratic Inverse Problemscited by 37
- 2019First-order methods almost always avoid saddle points: the case of vanishing step-sizescited by 35
- 2016Gradient Descent Only Converges to Minimizers: Non-Isolated Critical Points and Invariant Regionscited by 34
- 2018Hessian Riemannian gradient flows in convex programmingcited by 26
- 2023Adam-family Methods for Nonsmooth Optimization with Convergence Guaranteescited by 22
- 2022The effect of smooth parametrizations on nonconvex optimization landscapescited by 21
- 2021Fast Stochastic Bregman Gradient Methods: Sharp Analysis and Variance Reductioncited by 19
20 more
- 2019Optimal Complexity and Certification of Bregman First-Order Methodscited by 16
- 2019An Inertial Newton Algorithm for Deep Learningcited by 14
- 2018Hessian barrier algorithms for linearly constrained optimization problemscited by 8
- 2021Bregman Proximal Point Algorithm Revisited: A New Inexact Version and its Inertial Variantcited by 8
- 2023Nonsmooth nonconvex stochastic heavy ballcited by 8
- 2023Riemannian stochastic optimization methods avoid strict saddle pointscited by 8
- 2019An exact penalty approach for optimization with nonnegative orthogonality constraintscited by 7
- 2023Nonconvex Stochastic Bregman Proximal Gradient Method with Application to Deep Learningcited by 7
- 2023Global stability of first-order methods for coercive tame functionscited by 6
- 2024Stochastic Bregman Proximal Gradient Method Revisited: Kernel Conditioning and Painless Variance Reductioncited by 6
- 2020An efficient implementable inexact entropic proximal point algorithm for a class of linear programming problemscited by 5
- 2021Hessian barrier algorithms for non-convex conic optimizationcited by 5
- 2023Adam-family Methods with Decoupled Weight Decay in Deep Learningcited by 5
- 2024Stochastic Bregman Subgradient Methods for Nonsmooth Nonconvex Optimization Problemscited by 5
- 2025An Exact Penalty Approach for Equality Constrained Optimization over a Convex Setcited by 4
- 2023On Squared-Variable Formulationscited by 3
- 2019Bregman Proximal Framework for Deep Linear Neural Networkscited by 2
- 2024Developing Lagrangian-based Methods for Nonsmooth Nonconvex Optimizationcited by 2
- 2025A Bregman ADMM for Bethe variational problemcited by 2
- 2024Optimization over convex polyhedra via Hadamard parametrizationscited by 1
Built on by
Citing papers per year: 2025: 1; 2026: 2.
3 papers.