Trigger Optimization and Event Classification for Dark Matter Searches in the CYGNO Experiment Using Machine Learning

📅 2026-01-28
📈 Citations: 1
Influential: 0
📄 PDF
🤖 AI Summary
This work addresses the challenges of efficient real-time triggering, compression, and background suppression in high-resolution sparse optical imaging within the CYGNO experiment. To this end, two novel approaches are proposed: first, an unsupervised online compression framework based on a convolutional autoencoder that extracts regions of interest (ROIs) via reconstruction residuals, achieving fully unsupervised real-time ROI identification for the first time in CYGNO—retaining 93.0% of signal intensity while discarding 97.8% of background pixels with only 25 ms inference latency; second, the application of the weakly supervised Classification Without Labels (CWoLa) method to mixed data, which successfully identifies compact, circular nuclear recoil events and achieves performance approaching the theoretical optimum.

Technology Category

Application Category

📝 Abstract
The CYGNO experiment employs an optical-readout Time Projection Chamber (TPC) to search for rare low-energy interactions using finely resolved scintillation images. While the optical readout provides rich topological information, it produces large, sparse megapixel images that challenge real-time triggering, data reduction, and background discrimination. We summarize two complementary machine-learning approaches developed within CYGNO. First, we present a fast and fully unsupervised strategy for online data reduction based on reconstruction-based anomaly detection. A convolutional autoencoder trained exclusively on pedestal images (i.e. frames acquired with GEM amplification disabled) learns the detector noise morphology and highlights particle-induced structures through localized reconstruction residuals, from which compact Regions of Interest (ROIs) are extracted. On real prototype data, the selected configuration retains (93.0 +/- 0.2)% of reconstructed signal intensity while discarding (97.8 +/- 0.1)% of the image area, with ~25 ms per-frame inference time on a consumer GPU. Second, we report a weakly supervised application of the Classification Without Labels (CWoLa) framework to data acquired with an Americium--Beryllium neutron source. Using only mixed AmBe and standard datasets (no event-level labels), a convolutional classifier learns to identify nuclear-recoil-like topologies. The achieved performance approaches the theoretical limit imposed by the mixture composition and isolates a high-score population with compact, approximately circular morphologies consistent with nuclear recoils.
Problem

Research questions and friction points this paper is trying to address.

Dark Matter Search
Optical-readout TPC
Real-time Triggering
Data Reduction
Background Discrimination
Innovation

Methods, ideas, or system contributions that make the work stand out.

unsupervised anomaly detection
convolutional autoencoder
Classification Without Labels (CWoLa)
optical-readout TPC
real-time data reduction
🔎 Similar Papers
No similar papers found.
💼 Related Jobs
No related jobs found.
F
F. D. Amaro
LIBPhys, Department of Physics, University of Coimbra, 3004-516 Coimbra, Portugal
R
R. Antonietti
Dipartimento di Matematica e Fisica, Università Roma Tre, 00146 Roma, Italy; INFN Sezione di Roma Tre, 00146 Roma, Italy
E
E. Baracchini
Gran Sasso Science Institute, 67100 L’Aquila, Italy; INFN Laboratori Nazionali del Gran Sasso, 67100 Assergi, Italy
L
L. Benussi
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
C
C. Capoccia
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
M
M. Caponero
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy; ENEA Centro Ricerche Frascati, 00044 Frascati, Italy
L
L. G. M. de Carvalho
Universidade Federal de Juiz de Fora, Faculdade de Engenharia, 36036-900 Juiz de Fora, MG, Brazil
G
G. Cavoto
Dipartimento di Fisica, Sapienza Università di Roma, 00185 Roma, Italy; INFN Sezione di Roma, 00185 Roma, Italy
I
I. A. Costa
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
A
A. Croce
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
M
M. D'Astolfo
Gran Sasso Science Institute, 67100 L’Aquila, Italy; INFN Laboratori Nazionali del Gran Sasso, 67100 Assergi, Italy
G
G. D’Imperio
Dipartimento di Fisica, Sapienza Università di Roma, 00185 Roma, Italy; INFN Sezione di Roma, 00185 Roma, Italy
G
G. Dho
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
E
E. D. Marco
Dipartimento di Fisica, Sapienza Università di Roma, 00185 Roma, Italy; INFN Sezione di Roma, 00185 Roma, Italy
J
J. M. F. D. Santos
LIBPhys, Department of Physics, University of Coimbra, 3004-516 Coimbra, Portugal
D
D. Fiorina
Gran Sasso Science Institute, 67100 L’Aquila, Italy; INFN Laboratori Nazionali del Gran Sasso, 67100 Assergi, Italy
F
F. Iacoangeli
Dipartimento di Fisica, Sapienza Università di Roma, 00185 Roma, Italy; INFN Sezione di Roma, 00185 Roma, Italy
Z
Z. Islam
Gran Sasso Science Institute, 67100 L’Aquila, Italy; INFN Laboratori Nazionali del Gran Sasso, 67100 Assergi, Italy
E
E. Kemp
Universidade Estadual de Campinas (UNICAMP), Campinas 13083-859, SP, Brazil
H
H. P. Lima
Gran Sasso Science Institute, 67100 L’Aquila, Italy; INFN Laboratori Nazionali del Gran Sasso, 67100 Assergi, Italy
G
G. Maccarrone
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
R
R. Mano
LIBPhys, Department of Physics, University of Coimbra, 3004-516 Coimbra, Portugal
D
D. Marques
Gran Sasso Science Institute, 67100 L’Aquila, Italy; INFN Laboratori Nazionali del Gran Sasso, 67100 Assergi, Italy
G
G. Mazzitelli
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
P
P. Meloni
Dipartimento di Matematica e Fisica, Università Roma Tre, 00146 Roma, Italy; INFN Sezione di Roma Tre, 00146 Roma, Italy
A
A. Messina
Dipartimento di Fisica, Sapienza Università di Roma, 00185 Roma, Italy; INFN Sezione di Roma, 00185 Roma, Italy
C
C. Monteiro
LIBPhys, Department of Physics, University of Coimbra, 3004-516 Coimbra, Portugal
R
R. Nobrega
Universidade Federal de Juiz de Fora, Faculdade de Engenharia, 36036-900 Juiz de Fora, MG, Brazil
G
G. M. Oppedisano
Gran Sasso Science Institute, 67100 L’Aquila, Italy; INFN Laboratori Nazionali del Gran Sasso, 67100 Assergi, Italy
I
I. F. Pains
Universidade Federal de Juiz de Fora, Faculdade de Engenharia, 36036-900 Juiz de Fora, MG, Brazil
E
E. Paoletti
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
F
F. Petrucci
Dipartimento di Matematica e Fisica, Università Roma Tre, 00146 Roma, Italy; INFN Sezione di Roma Tre, 00146 Roma, Italy
S
S. Piacentini
Gran Sasso Science Institute, 67100 L’Aquila, Italy; INFN Laboratori Nazionali del Gran Sasso, 67100 Assergi, Italy
D
D. Pierluigi
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
D
D. Pinci
Dipartimento di Fisica, Sapienza Università di Roma, 00185 Roma, Italy; INFN Sezione di Roma, 00185 Roma, Italy
F
F. Renga
Dipartimento di Fisica, Sapienza Università di Roma, 00185 Roma, Italy; INFN Sezione di Roma, 00185 Roma, Italy
A
A. Russo
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
G
G. Saviano
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy; Dipartimento di Ingegneria Chimica, Materiali e Ambiente, Sapienza Università di Roma, 00185 Roma, Italy
P
P. A. O. C. Silva
LIBPhys, Department of Physics, University of Coimbra, 3004-516 Coimbra, Portugal
N
N. Spooner
Department of Physics and Astronomy, University of Sheffield, Sheffield S3 7RH, UK
R
R. Tesauro
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
S
S. Tomassini
INFN Laboratori Nazionali di Frascati, 00044 Frascati, Italy
D
D. Tozzi
Dipartimento di Fisica, Sapienza Università di Roma, 00185 Roma, Italy; INFN Sezione di Roma, 00185 Roma, Italy