{"id":"arxiv/2610.26748","url":"https://arxiv.org/abs/2610.26748","tags":["math.OC","cs.LG"],"score":0,"title":"Gradient descent reaches arbitrary accuracy in a fixed number of large steps on multiclass logistic regression","domain":"math.OC","source":"arxiv","authors":["Andreea-Alexandra Musat","Nicolas Boumal"],"reviews":0,"subject":"math.OC","summary":"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…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-19","reactions":{}}