Compressive Clustering with an Optical Processing Unit

TL;DR

This paper investigates using Optical Processing Units to efficiently compute random Fourier features for compressive clustering, introducing tools for hyper-parameter tuning to improve clustering performance.

Contribution

It adapts the compressive clustering pipeline to optical hardware and proposes methods for tuning key hyper-parameters.

Findings

Optical Processing Units can effectively compute random Fourier features.

The proposed tuning tools improve clustering results.

The approach offers a faster, hardware-accelerated clustering method.

Abstract

We explore the use of Optical Processing Units (OPU) to compute random Fourier features for sketching, and adapt the overall compressive clustering pipeline to this setting. We also propose some tools to help tuning a critical hyper-parameter of compressive clustering.