Toward a Unified Theory of Gradient Descent under Generalized Smoothness

TL;DR

This paper extends the analysis of gradient descent to functions with generalized smoothness, deriving adaptive step sizes and improving convergence guarantees in both convex and nonconvex optimization.

Contribution

It introduces a unified framework for gradient descent under generalized smoothness, deriving new step size rules and convergence results beyond classical L-smoothness assumptions.

Findings

Derived adaptive step size based on generalized smoothness.

Improved convergence rates over existing methods.

Extended analysis to new optimization settings.

Abstract

We study the classical optimization problem $\min_{x \in \mathbb{R}^d} f(x)$ and analyze the gradient descent (GD) method in both nonconvex and convex settings. It is well-known that, under the $L$-smoothness assumption ($\|\nabla^2 f(x)\| \leq L$), the optimal point minimizing the quadratic upper bound $f(x_k) + \langle\nabla f(x_k), x_{k+1} - x_k\rangle + \frac{L}{2} \|x_{k+1} - x_k\|^2$ is $x_{k+1} = x_k - \gamma_k \nabla f(x_k)$ with step size $\gamma_k = \frac{1}{L}$. Surprisingly, a similar result can be derived under the $\ell$-generalized smoothness assumption ($\|\nabla^2 f(x)\| \leq \ell(\|\nabla f(x)\|)$). In this case, we derive the step size $$\gamma_k = \int_{0}^{1} \frac{d v}{\ell(\|\nabla f(x_k)\| + \|\nabla f(x_k)\| v)}.$$ Using this step size rule, we improve upon existing theoretical convergence rates and obtain new results in several previously unexplored setups.