Answers are grounded in this paper and cite their sources.
We consider gradient descent (GD) with a constant stepsize applied to logistic regression with linearly separable data, where the constant stepsize η is so large that the loss initially oscillates. We show that GD exits this initial oscillatory phase rapidly -- in O(η) steps -- and subsequently achieves an Õ(1 / (η t) ) convergence rate after t additional steps. Our results imply that, given a budget of T steps, GD can achieve an accelerated loss of Õ(1/T²) with an aggressive stepsize η:= Θ( T), without any use of momentum or variable stepsize schedulers. Our proof technique is versatile and also handles general classification loss functions (where exponential tails are needed for the Õ(1/T²) acceleration), nonlinear predictors in the neural tangent kernel regime, and online stochastic gradient descent (SGD) with a large stepsize, under suitable separability conditions.
No takes yet. Share an insight, caveat, or question.
Wu et al. (2024) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: