Variational autoencoders for at-source data reduction and anomaly detection in high energy particle detectors

Detectors in next-generation high-energy physics experiments face several daunting requirements, such as high data rates, damaging radiation exposure, and stringent constraints on power, space, and latency. To address these challenges, machine learning in readout electronics can be leveraged for sma...

Full description

Saved in:
Bibliographic Details
Main Authors: Alexander Yue, Haoyi Jia, Julia Gonski
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:Machine Learning: Science and Technology
Subjects:
Online Access:https://doi.org/10.1088/2632-2153/adf0c0
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849245688616976384
author Alexander Yue
Haoyi Jia
Julia Gonski
author_facet Alexander Yue
Haoyi Jia
Julia Gonski
author_sort Alexander Yue
collection DOAJ
description Detectors in next-generation high-energy physics experiments face several daunting requirements, such as high data rates, damaging radiation exposure, and stringent constraints on power, space, and latency. To address these challenges, machine learning in readout electronics can be leveraged for smart detector designs, enabling intelligent inference and data reduction at-source. Variational autoencoders (VAEs) offer a variety of benefits for front-end readout; an on-sensor encoder can perform efficient lossy data compression while simultaneously providing a latent space representation that can be used for anomaly detection. Results are presented from low-latency and resource-efficient VAEs for front-end data processing in a futuristic silicon pixel detector. Encoder-based data compression is found to preserve good performance of off-detector analysis while significantly reducing the off-detector data rate as compared to a similarly sized data filtering approach. Furthermore, the latent space information is found to be a useful discriminator in the context of real-time sensor defect monitoring. Together, these results highlight the multifaceted utility of autoencoder-based front-end readout schemes and motivate their consideration in future detector designs.
format Article
id doaj-art-b5f8e0d72057446ab0fbc6abaee9c8b4
institution Kabale University
issn 2632-2153
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series Machine Learning: Science and Technology
spelling doaj-art-b5f8e0d72057446ab0fbc6abaee9c8b42025-08-20T03:58:44ZengIOP PublishingMachine Learning: Science and Technology2632-21532025-01-016303501710.1088/2632-2153/adf0c0Variational autoencoders for at-source data reduction and anomaly detection in high energy particle detectorsAlexander Yue0https://orcid.org/0000-0001-9901-9649Haoyi Jia1https://orcid.org/0000-0001-9191-3822Julia Gonski2https://orcid.org/0000-0003-2037-6315Stanford University , 450 Jane Stanford Way, Stanford, CA 94305, United States of AmericaStanford University , 450 Jane Stanford Way, Stanford, CA 94305, United States of America; SLAC National Accelerator Laboratory , 2575 Sand Hill Road MS 95, Menlo Park, CA 94025, United States of AmericaSLAC National Accelerator Laboratory , 2575 Sand Hill Road MS 95, Menlo Park, CA 94025, United States of AmericaDetectors in next-generation high-energy physics experiments face several daunting requirements, such as high data rates, damaging radiation exposure, and stringent constraints on power, space, and latency. To address these challenges, machine learning in readout electronics can be leveraged for smart detector designs, enabling intelligent inference and data reduction at-source. Variational autoencoders (VAEs) offer a variety of benefits for front-end readout; an on-sensor encoder can perform efficient lossy data compression while simultaneously providing a latent space representation that can be used for anomaly detection. Results are presented from low-latency and resource-efficient VAEs for front-end data processing in a futuristic silicon pixel detector. Encoder-based data compression is found to preserve good performance of off-detector analysis while significantly reducing the off-detector data rate as compared to a similarly sized data filtering approach. Furthermore, the latent space information is found to be a useful discriminator in the context of real-time sensor defect monitoring. Together, these results highlight the multifaceted utility of autoencoder-based front-end readout schemes and motivate their consideration in future detector designs.https://doi.org/10.1088/2632-2153/adf0c0machine learningedgeautoencodercolliderreadout electronics
spellingShingle Alexander Yue
Haoyi Jia
Julia Gonski
Variational autoencoders for at-source data reduction and anomaly detection in high energy particle detectors
Machine Learning: Science and Technology
machine learning
edge
autoencoder
collider
readout electronics
title Variational autoencoders for at-source data reduction and anomaly detection in high energy particle detectors
title_full Variational autoencoders for at-source data reduction and anomaly detection in high energy particle detectors
title_fullStr Variational autoencoders for at-source data reduction and anomaly detection in high energy particle detectors
title_full_unstemmed Variational autoencoders for at-source data reduction and anomaly detection in high energy particle detectors
title_short Variational autoencoders for at-source data reduction and anomaly detection in high energy particle detectors
title_sort variational autoencoders for at source data reduction and anomaly detection in high energy particle detectors
topic machine learning
edge
autoencoder
collider
readout electronics
url https://doi.org/10.1088/2632-2153/adf0c0
work_keys_str_mv AT alexanderyue variationalautoencodersforatsourcedatareductionandanomalydetectioninhighenergyparticledetectors
AT haoyijia variationalautoencodersforatsourcedatareductionandanomalydetectioninhighenergyparticledetectors
AT juliagonski variationalautoencodersforatsourcedatareductionandanomalydetectioninhighenergyparticledetectors