English

Central limit theorems for stochastic gradient descent with averaging for stable manifolds

Probability 2019-12-20 v1 Statistics Theory Statistics Theory

Abstract

In this article we establish new central limit theorems for Ruppert-Polyak averaged stochastic gradient descent schemes. Compared to previous work we do not assume that convergence occurs to an isolated attractor but instead allow convergence to a stable manifold. On the stable manifold the target function is constant and the oscillations in the tangential direction may be significantly larger than the ones in the normal direction. As we show, one still recovers a central limit theorem with the same rates as in the case of isolated attractors. Here we consider step-sizes γn=nγ\gamma_n=n^{-\gamma} with γ(34,1)\gamma\in(\frac34,1), typically.

Keywords

Cite

@article{arxiv.1912.09187,
  title  = {Central limit theorems for stochastic gradient descent with averaging for stable manifolds},
  author = {Steffen Dereich and Sebastian Kassing},
  journal= {arXiv preprint arXiv:1912.09187},
  year   = {2019}
}

Comments

42 pages

R2 v1 2026-06-23T12:50:58.971Z