Tensor Network-Constrained Kernel Machines as Gaussian Processes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wesel, Frederiek, Batselier, Kim
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916182842408960
author Wesel, Frederiek
Batselier, Kim
author_facet Wesel, Frederiek
Batselier, Kim
contents Tensor Networks (TNs) have recently been used to speed up kernel machines by constraining the model weights, yielding exponential computational and storage savings. In this paper we prove that the outputs of Canonical Polyadic Decomposition (CPD) and Tensor Train (TT)-constrained kernel machines recover a Gaussian Process (GP), which we fully characterize, when placing i.i.d. priors over their parameters. We analyze the convergence of both CPD and TT-constrained models, and show how TT yields models exhibiting more GP behavior compared to CPD, for the same number of model parameters. We empirically observe this behavior in two numerical experiments where we respectively analyze the convergence to the GP and the performance at prediction. We thereby establish a connection between TN-constrained kernel machines and GPs.
format Preprint
id arxiv_https___arxiv_org_abs_2403_19500
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Tensor Network-Constrained Kernel Machines as Gaussian Processes
Wesel, Frederiek
Batselier, Kim
Machine Learning
I.5.0
Tensor Networks (TNs) have recently been used to speed up kernel machines by constraining the model weights, yielding exponential computational and storage savings. In this paper we prove that the outputs of Canonical Polyadic Decomposition (CPD) and Tensor Train (TT)-constrained kernel machines recover a Gaussian Process (GP), which we fully characterize, when placing i.i.d. priors over their parameters. We analyze the convergence of both CPD and TT-constrained models, and show how TT yields models exhibiting more GP behavior compared to CPD, for the same number of model parameters. We empirically observe this behavior in two numerical experiments where we respectively analyze the convergence to the GP and the performance at prediction. We thereby establish a connection between TN-constrained kernel machines and GPs.
title Tensor Network-Constrained Kernel Machines as Gaussian Processes
topic Machine Learning
I.5.0
url https://arxiv.org/abs/2403.19500