End-to-end Kernel Learning via Generative Random Fourier Features

TL;DR

This paper introduces an end-to-end kernel learning framework using generative RFFs that jointly learns the kernel and classifier, leading to improved generalization and robustness in classification tasks.

Contribution

It proposes a novel one-stage, end-to-end approach combining kernel learning and classification via a generative network, surpassing traditional two-stage RFF methods.

Findings

Outperforms classical two-stage RFF methods in real-world classification tasks.

Enhances adversarial robustness through randomized resampling mechanisms.

Demonstrates superior generalization performance with deeper feature representations.

Abstract

Random Fourier features (RFFs) provide a promising way for kernel learning in a spectral case. Current RFFs-based kernel learning methods usually work in a two-stage way. In the first-stage process, learning the optimal feature map is often formulated as a target alignment problem, which aims to align the learned kernel with the pre-defined target kernel (usually the ideal kernel). In the second-stage process, a linear learner is conducted with respect to the mapped random features. Nevertheless, the pre-defined kernel in target alignment is not necessarily optimal for the generalization of the linear learner. Instead, in this paper, we consider a one-stage process that incorporates the kernel learning and linear learner into a unifying framework. To be specific, a generative network via RFFs is devised to implicitly learn the kernel, followed by a linear classifier parameterized as a full-connected layer. Then the generative network and the classifier are jointly trained by solving the empirical risk minimization (ERM) problem to reach a one-stage solution. This end-to-end scheme naturally allows deeper features, in correspondence to a multi-layer structure, and shows superior generalization performance over the classical two-stage, RFFs-based methods in real-world classification tasks. Moreover, inspired by the randomized resampling mechanism of the proposed method, its enhanced adversarial robustness is investigated and experimentally verified.