Newest papers

Gradient descent reaches arbitrary accuracy in a fixed number of large steps on multiclass logistic regression

Andreea-Alexandra Muşat, Nicolas Boumal · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26748 · cited by 0 works (OpenAlex) · open-access PDF

Spanish

Mostramos que el descenso gradiente (GD) con el tamaño de paso $η$ puede alcanzar $\varepsilon$ de la diferencia de optimidad en la pérdida de regresión logística multiclase en un número de iteraciones $T$ que es independiente de $\varepsilon$. Para ello, el tamaño de paso $η$ debe ser elegido lo suficientemente grande como una función de $\varepsilon$ (y propiedades del conjunto de datos linealmente separables), mientras que $T$ es una constante que depende sólo de los datos. Las propiedades relevantes de los datos incluyen el número de puntos y su margen (para $T$), y una cantidad adicional (para $η$) que resulta ser necesario en la dimensión $\dgeq 3$. Las pruebas se basan en el hecho conocido, que como $η\

Machine translation (Meta M2M100); the English original follows.

Abstract

We show that gradient descent (GD) with step size $η$ can achieve $\varepsilon$ optimality gap on the multiclass logistic regression loss in a number of iterations $T$ that is independent of $\varepsilon$. To do so, the step size $η$ must be chosen large enough as a function of $\varepsilon$ (and properties of the linearly separable dataset), while $T$ is a constant that only depends on the data. The relevant properties of the data include the number of points and their margin (for $T$), and one additional quantity (for $η$) that turns out to be necessary in dimension $d \geq 3$. The proofs build on the known fact that, as $η\to \infty$, suitably rescaled GD becomes the perceptron algorithm, and the latter finds a perfect classifier in $T$ iterations. A similar claim was shown for two classes in dimension $d = 2$ using other proof techniques. This result adds to previous works that showed GD on the logistic regression loss can be accelerated by taking large step sizes. Those provoke a wild, oscillatory initial phase, followed by a stable phase. We provide results of this (qualitatively different) nature, too.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Mathematics, Gradient descent, Separable space, Dimension (graph theory), Mathematical proof, Logistic regression

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange