{"id":"arxiv/2603.12209","url":"https://arxiv.org/abs/2603.12209","tags":["math.OC"],"score":0,"title":"Dictionary-Restricted First-Order Descent Methods: Bounds and Convergence Rates","domain":"math.OC","source":"arxiv","authors":["Miguel Berasategui","Pablo M. Berná","Antonio Falcó"],"reviews":0,"subject":"math.OC","summary":"This paper develops a general theory for first-order descent methods whose search directions are restricted to a prescribed dictionary in a reflexive Banach space. Instead of assuming that the linear span of the dictionary is dense, as in the classical Proper Generalized Decomposition framework of Falcó and Nouy or in the universality approach of Berná and Falcó, we introduce a geometric condition based on norming sets that guarantees density through a duality argument. This makes it possible to treat dictionaries arising from tensor formats, neural network units, and other nonlinear or…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-12","reactions":{}}