Exploiting the latent space of deep AutoEncoders for the identification of signal pulses in noisy time-series
Abstract
We propose a data-driven procedure, based on convolutional variational autoencoders, to identify the presence of signal pulses in long time-series. The dataset consists of synthetic waveforms, each composed of non-gaussian noise and a log-normal shaped signal of variable intensity, with a length of 10,000 samples. The model heavily compresses the input waveforms, allowing a direct study of such a reduced representation. After training for 150 epochs on 7,500 waveforms, a region in the latent space where the network encodes time-series presenting only background noise emerges, allowing in turn to tag as candidates for containing a signal those falling outside. When applied on a test dataset of freshly generated waveforms, 100% of the events with a large pulses are correctly labelled, and this fraction only decreases for signal amplitudes comparable with accidental noise pulses. This approach was designed to fully exploit the measurements in dual-phase Liquid Argon Time Projection Chambers, as the one of the Recoil Directionality experiment, built in the context of the Darkside project.
Versions
Related
- Physically Organized Latent Spaces in Unsupervised Autoencoders: Evidence from Aerodynamic Databases
- Reservoir computing generalized: Exploiting complex dynamics via time-invariant transformation
- Imaging-based visible light positioning exploiting yin-yang-yao-pattern fiducial markers on edge-lit LED panel luminaires
- Elucidating structure-property correlations in ferroelectric Hf0.5Zr0.5O2 films using variational autoencoders
- Pulses of Physics at the Heart of History
BibTeX
@misc{anastasi2026exploiting,
title = {{Exploiting the latent space of deep AutoEncoders for the identification of signal pulses in noisy time-series}},
author = {Gioacchino Alex Anastasi and Sebastiano Francesco Albergo and Marzio De Napoli and Noemi Pino and Sebastiana Maria Puglia and Alessia Rita Tricomi},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2607.12632},
eprint = {2607.12632},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.12632},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Exploiting the latent space of deep AutoEncoders for the identification of signal pulses in noisy time-series
AU - Gioacchino Alex Anastasi
AU - Sebastiano Francesco Albergo
AU - Marzio De Napoli
AU - Noemi Pino
AU - Sebastiana Maria Puglia
AU - Alessia Rita Tricomi
PY - 2026
DA - 2026/10/07
PB - arXiv
DO - 10.48550/arxiv.2607.12632
UR - https://arxiv.org/abs/2607.12632
AB - We propose a data-driven procedure, based on convolutional variational autoencoders, to identify the presence of signal pulses in long time-series. The dataset consists of synthetic waveforms, each composed of non-gaussian noise and a log-normal shaped signal of variable intensity, with a length of 10,000 samples. The model heavily compresses the input waveforms, allowing a direct study of such a reduced representation. After training for 150 epochs on 7,500 waveforms, a region in the latent space where the network encodes time-series presenting only background noise emerges, allowing in turn to tag as candidates for containing a signal those falling outside. When applied on a test dataset of freshly generated waveforms, 100% of the events with a large pulses are correctly labelled, and this fraction only decreases for signal amplitudes comparable with accidental noise pulses. This approach was designed to fully exploit the measurements in dual-phase Liquid Argon Time Projection Chambers, as the one of the Recoil Directionality experiment, built in the context of the Darkside project.
ER -