Pith. sign in

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

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
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.

citation-role summary

background 1

citation-polarity summary

fields

stat.ML 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

unclear 1

representative citing papers

Dimension-adapted Momentum Outscales SGD

stat.ML · 2025-05-22 · conditional · novelty 7.0

DANA, with dimension- and time-dependent momentum, provably outscales SGD on power-law random features when 2α>1, improving loss exponents and compute-optimal curves.

citing papers explorer

Showing 1 of 1 citing paper.

  • Dimension-adapted Momentum Outscales SGD stat.ML · 2025-05-22 · conditional · none · ref 62 · internal anchor

    DANA, with dimension- and time-dependent momentum, provably outscales SGD on power-law random features when 2α>1, improving loss exponents and compute-optimal curves.