Spectral Truncation Kernels: Noncommutativity in $C^*$-algebraic Kernel Machines

TL;DR

This paper introduces a new class of $C^*$-algebra-valued kernels based on spectral truncation, which incorporate noncommutativity to enhance the modeling of interactions in vector-valued data, bridging gaps in existing kernels.

Contribution

It proposes spectral truncation kernels that account for noncommutativity, providing a more flexible framework for vector-valued kernel learning and addressing kernel choice and computational challenges.

Findings

Fill the gap between separable and commutative kernels

Enable modeling of noncommutative interactions in data

Offer a flexible approach for vector-valued RKHSs

Abstract

$C^*$-algebra-valued kernels could pave the way for the next generation of kernel machines. To further our fundamental understanding of learning with $C^*$-algebraic kernels, we propose a new class of positive definite kernels based on the spectral truncation. We focus on kernels whose inputs and outputs are vectors or functions and generalize typical kernels by introducing the noncommutativity of the products appearing in the kernels. The noncommutativity induces interactions along the data function domain. We show that the proposed kernels fill the gap between existing separable and commutative kernels. We also propose a deep learning perspective to obtain a more flexible framework. The flexibility of the proposed class of kernels allows us to go beyond previous separable and commutative kernels, addressing two of the foremost issues regarding learning in vector-valued RKHSs, namely the choice of the kernel and the computational cost.