First-order algorithms converge faster than $O(1/k)$ on convex problems

Abstract

It is well known that both gradient descent and stochastic coordinate descent achieve a global convergence rate of $O(1/k)$ in the objective value, when applied to a scheme for minimizing a Lipschitz-continuously differentiable, unconstrained convex function. In this work, we improve this rate to $o(1/k)$. We extend the result to proximal gradient and proximal coordinate descent on regularized problems to show similar $o(1/k)$ convergence rates. The result is tight in the sense that a rate of $O(1/k^{1+ε})$ is not generally attainable for any $ε>0$, for any of these methods.

In the proceedings of the 36th International Conference on Machine Learning