Boosting Photon-Number-Resolved Detection Rates of Transition-Edge Sensors by Machine Learning

TL;DR

This paper introduces machine learning algorithms that significantly increase the detection rate of photon-number-resolving transition-edge sensors, enabling faster quantum photonic measurements without sacrificing accuracy.

Contribution

The authors develop supervised and unsupervised machine learning methods that enhance TES detection rates to 800 kHz while maintaining accurate photon-number resolution up to five photons.

Findings

Extended TES operation rate to 800 kHz

Maintained photon-number resolution accuracy up to five photons

Achieved at least four-fold improvement in detection rate

Abstract

Transition-Edge Sensors (TESs) are very effective photon-number-resolving (PNR) detectors that have enabled many photonic quantum technologies. However, their relatively slow thermal recovery time severely limits their operation rate in experimental scenarios compared to leading non-PNR detectors. In this work, we develop an algorithmic approach that enables TESs to detect and accurately classify photon pulses without waiting for a full recovery time between detection events. We propose two machine-learning-based signal processing methods: one supervised learning method and one unsupervised clustering method. By benchmarking against data obtained using coherent states and squeezed states, we show that the methods extend the TES operation rate to 800 kHz, achieving at least a four-fold improvement, whilst maintaining accurate photon-number assignment up to at least five photons. Our algorithms will find utility in applications where high rates of PNR detection are required and in technologies which demand fast active feed-forward of PNR detection outcomes.