Quantum optical classifier with superexponential speedup

TL;DR

This paper introduces a quantum optical pattern recognition method that classifies objects with constant complexity, achieving a superexponential speedup over classical neural networks by encoding data into single-photon states and using quantum interference.

Contribution

It presents a novel quantum optical classifier that operates with constant complexity, surpassing classical neural networks in efficiency for binary classification tasks.

Findings

Constant $\\mathcal{O}(1)$ complexity in resources required

Superexponential speedup over classical neurons

Effective binary classification without image reconstruction

Abstract

Classification is a central task in deep learning algorithms. Usually, images are first captured and then processed by a sequence of operations, of which the artificial neuron represents one of the fundamental units. This paradigm requires significant resources that scale (at least) linearly in the image resolution, both in terms of photons and computational operations. Here, we present a quantum optical pattern recognition method for binary classification tasks. It classifies objects without reconstructing their images, using the rate of two-photon coincidences at the output of a Hong-Ou-Mandel interferometer, where both the input and the classifier parameters are encoded into single-photon states. Our method exhibits the behaviour of a classical neuron of unit depth. Once trained, it shows a constant $\mathcal{O}(1)$ complexity in the number of computational operations and photons required by a single classification. This is a superexponential advantage over a classical artificial neuron.