On the Complexity of Lower-Order Implementations of Higher-Order Methods

Abstract

In this work, we propose a method for minimizing non-convex functions with Lipschitz continuous $p$th-order derivatives, starting from $p\ge 1$. The method, however, only requires derivative information up to order $(p-1)$, since the $p$th-order derivatives are approximated via finite differences. To ensure oracle efficiency, instead of computing finite-difference approximations at every iteration, we reuse each approximation for $m$ consecutive iterations before recomputing it, with $m\ge 1$ as a key parameter. As a result, we obtain an adaptive method of order $(p-1)$ that requires no more than $O(\hat{I}{\mu }^{-\frac{p+1}{p}})$ iterations to find an $\hat{I}\mu$-approximate stationary point of the objective function and that, for the choice $m=(p-1)n+1$, where $n$ is the problem dimension, takes no more than $O(n^{1/p}\hat{I}{\mu }^{-\frac{p+1}{p}})$ oracle calls of order $(p-1)$. This improves previously known bounds for tensor methods with finite-difference approximations in terms of the problem dimension.