OCLGMLDec 3, 2024

Improved Complexity for Smooth Nonconvex Optimization: A Two-Level Online Learning Approach with Quasi-Newton Methods

arXiv:2412.02175v110 citationsh-index: 10STOC
Originality Incremental advance
AI Analysis

This work addresses optimization efficiency for nonconvex problems in machine learning, offering an incremental improvement in complexity bounds.

The paper tackles the problem of finding an ε-first-order stationary point for smooth nonconvex functions using gradient queries, achieving a gradient complexity of O(d^{1/4}ε^{-13/8}), which improves upon the best-known O(ε^{-7/4}) when d = O(ε^{-1/2}).

We study the problem of finding an $ε$-first-order stationary point (FOSP) of a smooth function, given access only to gradient information. The best-known gradient query complexity for this task, assuming both the gradient and Hessian of the objective function are Lipschitz continuous, is ${O}(ε^{-7/4})$. In this work, we propose a method with a gradient complexity of ${O}(d^{1/4}ε^{-13/8})$, where $d$ is the problem dimension, leading to an improved complexity when $d = {O}(ε^{-1/2})$. To achieve this result, we design an optimization algorithm that, underneath, involves solving two online learning problems. Specifically, we first reformulate the task of finding a stationary point for a nonconvex problem as minimizing the regret in an online convex optimization problem, where the loss is determined by the gradient of the objective function. Then, we introduce a novel optimistic quasi-Newton method to solve this online learning problem, with the Hessian approximation update itself framed as an online learning problem in the space of matrices. Beyond improving the complexity bound for achieving an $ε$-FOSP using a gradient oracle, our result provides the first guarantee suggesting that quasi-Newton methods can potentially outperform gradient descent-type methods in nonconvex settings.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes