Image de couverture de livre

First-order methods almost always avoid saddle points: the case of vanishing step-sizes

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Panageas, Ioannis, Piliouras, Georgios, Wang, Xiao
Format:	Preprint
Publié:	2019
Sujets:	Optimization and Control
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

An inexact primal-dual method with correction step for a saddle point problem in image debluring
par: Fang, Changjie, et autres
Publié: (2021)

The Limit Points of (Optimistic) Gradient Descent in Min-Max Optimization
par: Daskalakis, Constantinos, et autres
Publié: (2018)

Second-order methods for provably escaping strict saddle points in composite nonconvex and nonsmooth optimization
par: Bodard, Alexander, et autres
Publié: (2025)

First-order algorithms for robust optimization problems via convex-concave saddle-point Lagrangian reformulation
par: Postek, Krzysztof, et autres
Publié: (2021)

An inexact golden ratio primal-dual algorithm with linesearch step for a saddle point problem
par: Fang, Changjie, et autres
Publié: (2024)

Randomized coordinate gradient descent almost surely escapes strict saddle points
par: Chen, Ziang, et autres
Publié: (2025)

Last-Iterate Convergence: Zero-Sum Games and Constrained Min-Max Optimization
par: Daskalakis, Constantinos, et autres
Publié: (2018)

A stochastic preconditioned Douglas-Rachford splitting method for saddle-point problems
par: Dong, Yakun, et autres
Publié: (2022)

Gradient-free algorithm for saddle point problems under overparametrization
par: Statkevich, Ekaterina, et autres
Publié: (2024)

Avoiding strict saddle points of nonconvex regularized problems
par: Bai, Luwei, et autres
Publié: (2024)

Escaping saddle points without Lipschitz smoothness: the power of nonlinear preconditioning
par: Bodard, Alexander, et autres
Publié: (2025)

Proximal gradient descent on the smoothed duality gap to solve saddle point problems
par: Fercoq, Olivier
Publié: (2025)

Defining Lyapunov functions as the solution of a performance estimation saddle point problem
par: Fercoq, Olivier
Publié: (2024)

The Mirror-Prox Sliding Method for Non-smooth decentralized saddle-point problems
par: Kuruzov, Ilya, et autres
Publié: (2022)

Dealing with unbounded gradients in stochastic saddle-point optimization
par: Neu, Gergely, et autres
Publié: (2024)

Avoidance of non-strict saddle points by blow-up
par: Achour, El Mehdi, et autres
Publié: (2025)

Riemannian trust-region methods for strict saddle functions with complexity guarantees
par: Goyens, Florentin, et autres
Publié: (2024)

Exact Convergence rate of the subgradient method by using Polyak step size
par: Zamani, Moslem, et autres
Publié: (2024)

A two-step inertial method with a new step-size rule for variational inequalities in hilbert spaces
par: Peng, Jian-Wen, et autres
Publié: (2026)

Online Learning on Hidden-Convex Losses via Algorithmic Equivalence: Optimal Regret, Geometric Barrier, and Bandit Feedback
par: Barakat, Anas, et autres
Publié: (2026)

Shadow splitting methods for nonconvex optimisation: epi-approximation, convergence and saddle point avoidance
par: Atenas, Felipe
Publié: (2025)

Eigenvalue bounds for preconditioned symmetric multiple saddle-point matrices
par: Bergamaschi, L., et autres
Publié: (2025)

Adaptive primal dual hybrid gradient algorithms based on average spectrum for saddle point problems
par: Xu, Shengjie, et autres
Publié: (2026)

Non-ergodic convergence rate of an inertial accelerated primal-dual algorithm for saddle point problems
par: He, X., et autres
Publié: (2023)

Faster Rates for No-Regret Learning in General Games via Cautious Optimism
par: Soleymani, Ashkan, et autres
Publié: (2025)

Cautious Optimism: A Meta-Algorithm for Near-Constant Regret in General Games
par: Soleymani, Ashkan, et autres
Publié: (2025)

Data-Scarce Identification of Game Dynamics via Sum-of-Squares Optimization
par: Sakos, Iosif, et autres
Publié: (2023)

Exact penalty functions and global saddle points of augmented Lagrangians for well-posed constrained optimization problems
par: Dolgopolik, M. V.
Publié: (2025)

One Benders cut to rule all schedules in the neighbourhood
par: Avgerinos, Ioannis, et autres
Publié: (2023)

Resource-dependent process times in hybrid flexible flowshops
par: Avgerinos, Ioannis, et autres
Publié: (2025)

A convex combination based primal-dual algorithm with linesearch for general convex-concave saddle point problems
par: Chang, Xiaokai, et autres
Publié: (2024)

Contractivity and linear convergence in bilinear saddle-point problems: An operator-theoretic approach
par: Dirren, Colin, et autres
Publié: (2024)

Tailored First-order and Interior-point methods and a new semidefinite programming hierarchy for entanglement detection
par: Pena, Javier, et autres
Publié: (2025)

Online Multi-Agent Control with Adversarial Disturbances
par: Barakat, Anas, et autres
Publié: (2025)

Another approach to build Lyapunov functions for the first order methods in the quadratic case
par: Merkulov, Daniil, et autres
Publié: (2023)

A primal-dual interior point trust region method for second-order stationary points of Riemannian inequality-constrained optimization problems
par: Obara, Mitsuaki, et autres
Publié: (2025)

An adaptive linearized alternating direction multiplier method with a relaxation step for convex programming
par: Wang, Boran
Publié: (2024)

GeNIOS: an (almost) second-order operator-splitting solver for large-scale convex optimization
par: Diamandis, Theo, et autres
Publié: (2023)

Convergence rates of stochastic gradient method with independent sequences of step-size and momentum weight
par: Hwang, Wen-Liang
Publié: (2024)

First-order methods for stochastic and finite-sum convex optimization with deterministic constraints
par: Lu, Zhaosong, et autres
Publié: (2025)