{"ID":2869362,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.14969","arxiv_id":"2509.14969","title":"Stochastic Adaptive Gradient Descent Without Descent","abstract":"We introduce a new adaptive step-size strategy for convex optimization with stochastic gradient that exploits the local geometry of the objective function only by means of a first-order stochastic oracle and without any hyper-parameter tuning. The method comes from a theoretically-grounded adaptation of the Adaptive Gradient Descent Without Descent method to the stochastic setting. We prove the convergence of stochastic gradient descent with our step-size under various assumptions, and we show that it empirically competes against tuned baselines.","short_abstract":"We introduce a new adaptive step-size strategy for convex optimization with stochastic gradient that exploits the local geometry of the objective function only by means of a first-order stochastic oracle and without any hyper-parameter tuning. The method comes from a theoretically-grounded adaptation of the Adaptive Gr...","url_abs":"https://arxiv.org/abs/2509.14969","url_pdf":"https://arxiv.org/pdf/2509.14969v1","authors":"[\"Jean-François Aujol\",\"Jérémie Bigot\",\"Camille Castera\"]","published":"2025-09-18T14:02:10Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"math.OC\",\"stat.ML\"]","methods":"[]","has_code":false}