Work

The global convergence of stochastic gradient descent in non-convex landscapes: Sharp estimates via large deviations

Year: 2025

Type: conference-paper

Abstract: International audience

Source: HAL (Le Centre pour la Communication Scientifique Directe)

Authors: Azizian, Waïss, Iutzeler, Franck, Malick, Jérôme, Mertikopoulos, Panayotis

Institutions: Laboratoire d'Informatique de Grenoble, POLARIS: Évaluation et optimisation des performances des grandes infrastructures, Université Toulouse III - Paul Sabatier, Institut National des Sciences Appliquées de Toulouse, Institut de Mathématiques de Toulouse

Language: English

Topic: Stochastic Gradient Optimization Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: European Commission, Agence Nationale de la Recherche, NextGenerationEU

Awards: MIAI @ Grenoble Alpes, G2121614283, Foundations of robustness and reliability in artificial intelligence, G292453335, G3010527862, Systemes et Algorithmes Pervasifs au confluent des mondes physique et numérique, G6276162139, G994248709

OpenAlex ID: W7010090961

FWCI:

Cites:

Cited by:

Related to: