A Random Matrix Analysis of Random Fourier Features: Beyond the Gaussian Kernel, a Precise Phase Transition, and the Corresponding Double Descent

TL;DR

This paper provides a detailed random matrix analysis of random Fourier features in high-dimensional regimes, revealing phase transitions, double descent phenomena, and accurate error estimates without strong data assumptions.

Contribution

It offers the first precise characterization of RFF regression behavior in large, comparable data and feature dimensions, including phase transitions and double descent curves.

Findings

Identifies a phase transition in RFF regression performance.

Derives accurate training and test error estimates.

Matches theoretical predictions with real-world data experiments.

Abstract

This article characterizes the exact asymptotics of random Fourier feature (RFF) regression, in the realistic setting where the number of data samples $n$, their dimension $p$, and the dimension of feature space $N$ are all large and comparable. In this regime, the random RFF Gram matrix no longer converges to the well-known limiting Gaussian kernel matrix (as it does when $N \to \infty$ alone), but it still has a tractable behavior that is captured by our analysis. This analysis also provides accurate estimates of training and test regression errors for large $n,p,N$. Based on these estimates, a precise characterization of two qualitatively different phases of learning, including the phase transition between them, is provided; and the corresponding double descent test error curve is derived from this phase transition behavior. These results do not depend on strong assumptions on the data distribution, and they perfectly match empirical results on real-world data sets.