Machine Learning Approaches for Improved Scalability of Metallic Magnetic Calorimeters
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.
Kai: I'm Kai.
Mira: And I'm Mira, here with Lev.
Kai: We're looking at the paper: "Machine Learning Approaches for Improved Scalability of Metallic Magnetic Calorimeters".
Mira: Here's the gist of the paper. Metallic Magnetic Calorimeters (MMCs) are promising new tools for high precision X-ray spectroscopy, and this work explores applying Machine Learning (ML) methods to address challenges in signal processing and scalability The paper investigates the application of Machine Learning (ML) methods to improve the performance and scalability of Metallic Magnetic Calorimeters (MMCs) by addressing complexities in detector response and processing pipelines ML can be used for pulse classification, artifact rejection, pulse shape analysis, and feature extraction The authors demonstrate that ML can provide robust and scalable solutions for MMC signal processing by leveraging unsupervised learning for label auto-discovery and supervised learning for classification and regression tasks Pulse Classification and Artifact Rejection One example of ML application is the pulse classifier used to reject noise caused by external sources Traditional methods relying on simple thresholding are insufficient for capturing the full complexity of data and require extensive manual tuning The authors employ a Reduce-Cluster-Classify pipeline involving dimensionality reduction using PCA and UMAP to project the high-dimensional data into a lower space Subsequently, HDBSCAN is used for clustering to identify groups of similar patterns, where real pulses are expected to form the largest cluster A Random Forest Classifier is then trained on these auto-discovered labels to predict artifacts directly from raw traces Pulse Shape Analysis and Feature Extraction The second example focuses on extracting relevant features like pulse amplitude and trigger time, which traditionally relies on hand-optimized FIR filters ML methods, particularly Neural Networks (NN), learn the mapping from raw pulses to features directly from the data, allowing for automatic adaptation to different pulse shapes and noise characteristics This is achieved through a simulation-to-real domain adaptation approach where models are trained on simulated data and then fine-tuned with a small subset of real detector data A Convolutional Variational Autoencoder (CVAE) is used for unsupervised reconstruction, learning a compact latent space representation of the pulse shapes A Multi-Layer Perceptron (MLP) then maps this latent space representation to the relevant features using latent space regression Future Directions and Conclusion The paper concludes that ML methods offer significant advantages over traditional techniques by enhancing the existing processing pipeline without acting as an unexplained black box Future work will involve upgrading simulation data sources, incorporating more real-world data using GANs to bridge the gap between simulation and reality Time series analysis via LSTM networks could be investigated to model and correct for temporal drifts in the detector response Furthermore, a bespoke Bayesian optimization framework could target hardware optimization of SQUIDs based on real-time feedback Author Contributions M.O.
Kai: Herdrich is credited with Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Visualization and Writing – original draft T.
Mira: Mattis contributed to Conceptualization and Methodology validation G.
Kai: Weber provided Support and data provision for the evaluation of the presented methods D.A.
Paper summary: Mira: So, wrapping up this discussion on "Machine Learning Approaches for Improved Scalability of Metallic Magnetic Calorimeters," what’s left as the main message from M.O. Herdrich and their team?
Kai: The main point is that machine learning provides a way to enhance the existing processing pipeline for metallic magnetic calorimeters by addressing challenges in pulse classification and feature extraction2606 point 25045#pg1 <ref:2606.25045#pg1>. They demonstrated how ML can provide robust and scalable solutions for MMC signal processing by using unsupervised learning for label auto-discovery and supervised learning for classification and regression tasks2606 point 25045#pg1 <ref:2606.25045#pg1>.
Lev: It seems they are trying to make the processing more flexible, moving away from fixed heuristics that might not be sufficient as the detector parameters change over time3. That flexibility is what makes it scalable for future detector chips with thousands of pixels2606 point 25045#pg2 <ref:2606.25045#pg1>.
Mira: I see it as a way to manage the inherent difficulty of having no true values available for physical quantities in measured data, which is the largest challenge mentioned in their work3. The paper argues that while direct supervised learning is impossible without labels, ML can still provide useful predictions for us2606 point 25045#pg3 <ref:2606.25045#pg1>.
Kai: Exactly, so they’re not claiming a magic fix for everything because they acknowledge the difficulty with over fitting and spurious correlations too23. It's about building better tools to handle the inherent noise and complexity of these detectors in a scalable way2606 point 25045#pg1 <ref:2606.25045#pg1>.
Lev: For us, it means we need to think about how this feedback loop could be used to make the hardware optimization process less time consuming and more accurate through real-time data3.
Mira: Ultimately, the implication is that machine learning isn't replacing the physics; it’s enhancing how we analyze the experimental data generated by these high precision instruments2606 point 25045#pg1 <ref:2606.25045#pg1>.
Conclusion: Kai: So, to wrap up this session on Machine Learning Approaches for Improved Scalability of Metallic Magnetic Calorimeters, the core idea is that ML tools can make processing these calorimeters much more robust and scalable than what we have now.
Mira: Right, it’s about using machine learning for pulse classification and feature extraction so we don't have to rely on just simple thresholding or manually tuned filters.
Lev: Exactly, the authors are showing how unsupervised learning can help them discover labels automatically, while supervised learning predicts artifacts directly from the raw signal traces.
Kai: It feels like they’re building a smarter way to handle the noise and complexity inherent in these high-precision detectors without having to build an impossibly complex manual calibration routine for every single detector setup.
Mira: I see it as moving away from those fixed methods toward something that can adapt automatically when the detector response starts drifting, which is a real issue with long-term experiments.
Lev: From my side, if this works on real hardware, it means we could reduce the computational overhead for error correction and get better results in a much larger array of detectors.
Kai: So essentially, these ML methods are about making the existing experimental pipeline stronger without turning it into some completely opaque black box we can't debug.
Mira: That’s right, they aren't just adding complexity; they’re proposing enhancements to the existing processing flow that actually help with scalability and robustness in a way traditional physics modeling couldn't easily achieve alone.
Lev: But what this really means for us is figuring out how to integrate this kind of learning directly into the real-time control systems for future detector arrays.
M.O. Herdrich, T. Mattis, G. Weber, D.A. Schnaußl-Müller, J.H.Á Walch, Th.Ũ Stζhlker
GSI Helmholtzzentrum fеr Schwerionenforschung · Helmholtz-Institute Jena · Institute for Optics- and Quantumelectronics, Friedrich-Schiller-University Jena · Hasso-Plattner-Institute
physics.ins-det
Submitted: 2026-06-23
Updated: 2026-11-13
Comments: Has been accepted by the Journal for Applied Physics
Journal ref: M. O. Herdrich, T. Mattis, G. Weber, D. A. Schnauß-Müller, J. H. Walch, Th. Stöhlker; Machine learning approaches for improved scalability of metallic magnetic calorimeters. J. Appl. Phys. 7 November 2026; 140 (17): 174501
DOI: 10.1063/5.0348962
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 77/100
The gist: Metallic Magnetic Calorimeters (MMCs) are promising new tools for high precision X-ray spectroscopy, and this work explores applying Machine Learning (ML) methods to address challenges in signal…
Key concepts
- Pulse Classification
- ML is used to sort incoming signals into different categories, such as real pulses versus noise artifacts. Instead of simple thresholding, this method uses dimensionality reduction (PCA/UMAP) followed by clustering (HDBSCAN) to find patterns and then a Random Forest Classifier to predict the type of pulse.
- Pulse Shape Analysis
- This involves using Neural Networks (NN) to automatically extract important characteristics from raw pulse shapes, like amplitude and timing. A simulation-to-real adaptation approach, using a CVAE for reconstruction and an MLP for feature mapping, allows the model to learn features directly from data.
- Unsupervised Learning
- This technique is used to discover patterns in data without pre-labeled examples. In this study, unsupervised learning helps automatically find groups of similar pulse patterns (clusters) within the MMC signals, which helps in identifying what constitutes a 'real' pulse versus an artifact.
- Domain Adaptation
- This involves training ML models on simulated data and then fine-tuning them with a small set of real experimental data. This bridges the gap between perfect simulations and real-world detector behavior, ensuring the learned features are accurate for actual measurements.
Terminology
Summary
Metallic Magnetic Calorimeters (MMCs) are promising new tools for high precision X-ray spectroscopy, and this work explores applying Machine Learning (ML) methods to address challenges in signal processing and scalability
Introduction to ML Application
The paper investigates the application of Machine Learning (ML) methods to improve the performance and scalability of Metallic Magnetic Calorimeters (MMCs) by addressing complexities in detector response and processing pipelines ML can be used for pulse classification, artifact rejection, pulse shape analysis, and feature extraction The authors demonstrate that ML can provide robust and scalable solutions for MMC signal processing by leveraging unsupervised learning for label auto-discovery and supervised learning for classification and regression tasks
Pulse Classification and Artifact Rejection
One example of ML application is the pulse classifier used to reject noise caused by external sources Traditional methods relying on simple thresholding are insufficient for capturing the full complexity of data and require extensive manual tuning The authors employ a Reduce-Cluster-Classify pipeline involving dimensionality reduction using PCA and UMAP to project the high-dimensional data into a lower space Subsequently, HDBSCAN is used for clustering to identify groups of similar patterns, where real pulses are expected to form the largest cluster A Random Forest Classifier is then trained on these auto-discovered labels to predict artifacts directly from raw traces
Pulse Shape Analysis and Feature Extraction
The second example focuses on extracting relevant features like pulse amplitude and trigger time, which traditionally relies on hand-optimized FIR filters ML methods, particularly Neural Networks (NN), learn the mapping from raw pulses to features directly from the data, allowing for automatic adaptation to different pulse shapes and noise characteristics This is achieved through a simulation-to-real domain adaptation approach where models are trained on simulated data and then fine-tuned with a small subset of real detector data A Convolutional Variational Autoencoder (CVAE) is used for unsupervised reconstruction, learning a compact latent space representation of the pulse shapes A Multi-Layer Perceptron (MLP) then maps this latent space representation to the relevant features using latent space regression
Future Directions and Conclusion
The paper concludes that ML methods offer significant advantages over traditional techniques by enhancing the existing processing pipeline without acting as an unexplained black box Future work will involve upgrading simulation data sources, incorporating more real-world data using GANs to bridge the gap between simulation and reality Time series analysis via LSTM networks could be investigated to model and correct for temporal drifts in the detector response Furthermore, a bespoke Bayesian optimization framework could target hardware optimization of SQUIDs based on real-time feedback
Author Contributions
M.O. Herdrich is credited with Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Visualization and Writing – original draft T. p. 6 Author Contributions
M. Herdrich: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Visualization, Writing – original draft, Writing – review
and editing. Herdrich: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Visualization, Writing – original draft, Writing – review
and editing. Mattis: Conceptualization, Methodology, Validation, Writing – review and editing. Mattis: Conceptualization, Methodology, Validation, Writing – review and editing. Weber: Support and data provision for the evaluation of the
presented methods. Weber: Support and data provision for the evaluation of the
presented methods. Schnauß-Müller: Support and data provision for the
evaluation of the presented methods. Schnauß-Müller: Support and data provision for the
evaluation of the presented methods. Walch: Support and data provision for the evaluation of
the presented methods.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Acknowledgments
The authors acknowledge colleagues at the HI-Jena, the GSI, and the group of Prof. Dr. C. Enss at the Kirchhoff Institute for Physics for providing experimental data used for demonstration and evaluation purposes The publication is funded by the Open Access Publishing Fund of GSI Helmholtzzentrum fuer Schwerionenforschung
References
1M. Herdrich, Anwendung kryogener Kalorimeter für hoch aufgelöste Präzisions-Röntgenspektroskopie, Ph. p. 6 REFERENCES
1M. Herdrich, Anwendung kryogener Kalorimeter für hoch aufgelöste
Präzisions-Röntgenspektroskopie, Ph.Where the paper says it
2C. Enss, Cryogenic Particle Detection, Topics in Applied Physics (Springer-Verlag, Berlin Heidelberg, 2005)
3S. Kempf, A. Ferring, A. p. 6 Enss, Cryogenic Particle Detection, Topics in Applied Physics
(Springer-Verlag, Berlin Heidelberg, 2005). Ferring, A. Ferring, A. Fleischmann, and C.Where the paper says it
Where the paper says it
4J. Geist, Bestimmung Der Isomerenergie von 229Th Mit Dem Hochauflösenden Mikrokalorimeter-Array maXs30, Ph. p. 6 Enss, “Direct-current superconducting quantum interference devices for the readout of metallic magnetic calorimeters,” Superconductor Science and Technology 28, 045008
(2015). Geist, Bestimmung Der Isomerenergie von 229Th Mit Dem Hochauflösenden Mikrokalorimeter-Array maXs30, Ph.Where the paper says it
5C. Pies, S. Schäfer, S. p. 6 thesis, Ruprecht-KarlsUniversität Heidelberg, Heidelberg (2020). Schäfer, S. Schäfer, S. Pabinger, J. Pabinger, J. Ranitsch, N. Ranitsch, N. Foerster, D.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
6A. Fleischmann, C. p. 6 Ferring, A. Fleischmann, and C. Acknowledgments
The authors gratefully acknowledge the colleagues at the
HI-Jena, the GSI, and the group of Prof. Enss at the
Kirchhoff Institute for Physics, Heidelberg, Germany for providing experimental data used for demonstration and evaluation purposes in this study. Seidel, “Metallic Magnetic Calorimeters,” in Cryogenic Particle Detection, Topics in Applied Physics, edited by
C. Enss (Springer, Berlin, Heidelberg, 2005) pp.Where the paper says it
Where the paper says it
Where the paper says it
7P. Pfäfflein, G. p. 7 Machine Learning Approaches for Improved Scalability of Metallic Magnetic Calorimeters 8
7P. Pfäfflein, G. Machine Learning Approaches for Improved Scalability of Metallic Magnetic Calorimeters 8
7P. Pfäfflein, G.Where the paper says it
Where the paper says it
8M. O. Herdrich, D. p. 0 Machine Learning Approaches for Improved Scalability of Metallic Magnetic Calorimeters
Machine Learning Approaches for Improved Scalability of Metallic
Magnetic Calorimetersa)
M. Herdrich,1, 2, 3 T. Machine Learning Approaches for Improved Scalability of Metallic Magnetic Calorimeters
Machine Learning Approaches for Improved Scalability of Metallic
Magnetic Calorimetersa)
M. Herdrich,1, 2, 3 T. Machine Learning Approaches for Improved Scalability of Metallic Magnetic Calorimeters
Machine Learning Approaches for Improved Scalability of Metallic
Magnetic Calorimetersa)
M. Herdrich,1, 2, 3 T. Fleischmann, C. Gumberidze,
P.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
9M. O. Herdrich, D. Hengstler, M Keller J Geist C Schötz M Krantz A Fleischmann C Enss T Gassner P. p. 7 Hengstler, M. Schötz, M.Where the paper says it
10M. O. Herdrich, D. Hengstler, S Allgeier M Friedrich A Fleischmann C Enss S Bernitt T Morgenroth S Trotsenko R Schuch and T Stöhlker “Application of a metallic-magnetic calorimeter for high-resolution Xray spectroscopy of Fe at an EBIT” Journal of Physics B: Atomic, Molecular and Optical Physics (2024), 10. p. 7 Schuch, and T. Stöhlker, “Application of a metallic-magnetic calorimeter for high-resolution Xray spectroscopy of Fe at an EBIT,” Journal of Physics B: Atomic, Molecular and Optical Physics (2024), 10.Where the paper says it
11Ph. Pfäfflein, G. Weber, S Allgeier Z Andelkovic S Bernitt A I Bondarev A Borovik L Duval A Fleischmann O Forstner M Friedrich J Glorius A Gumberidze Ch Hahn F Herfurth D Hengstler M O Herdrich P.-M. Hillenbrand A Kalinin M Kiffer F M Kröger M Kubullek P Kuntz M Lestinsky Yu. p. 7 Kröger, M. Kubullek,
P.Where the paper says it
12D. Hengstler, Development and Characterization of Two-Dimensional Metallic Magnetic Calorimeter Arrays for the High-Resolution X-ray Spectroscopy, Ph. p. 7 Stöhlker, “Quantum Electrodynamics in
Strong Electromagnetic Fields: Substate Resolved K α Transition Energies
in Heliumlike Uranium,” Physical Review Letters 134, 153001 (2025). Hengstler, Development and Characterization of Two-Dimensional
Metallic Magnetic Calorimeter Arrays for the High-Resolution X-ray Spectroscopy, Ph.Where the paper says it
13A. Fleischmann, Magnetische Mikrokalorimeter: Hochauflösende Röntgenspektroskopie Mit Energiedispersiven Detektoren, Ph. p. 7 thesis, Ruprecht-Karls-Universität Heidelberg, Heidelberg
(2017). Fleischmann, Magnetische Mikrokalorimeter: Hochauflösende Röntgenspektroskopie Mit Energiedispersiven Detektoren, Ph.Where the paper says it
14C. Pies, maXs-200: Entwicklung Und Charakterisierung Eines Röntgendetektors Basierend Auf Magnetischen Kalorimetern Für Die Hochauflösende Spektroskopie Hochgeladener Ionen, Ph. p. 7 thesis,
Ruprecht-Karls-Universität Heidelberg, Heidelberg (2003). Pies, maXs-200: Entwicklung Und Charakterisierung Eines Röntgendetektors Basierend Auf Magnetischen Kalorimetern Für Die Hochauflösende Spektroskopie Hochgeladener Ionen, Ph.Where the paper says it
15M. Wegner, N. Karcher, O Krömer D Richter F Ahrens O Sander S Kempf M Weber and C Enss “Microwave SQUID Multiplexing of Metallic Magnetic Calorimeters: Status of Multiplexer Performance and Room-Temperature Readout Electronics Development” Journal of Low Temperature Physics 193, 462–475 (2018)
16C. M. Bishop, Pattern Recognition and Machine Learning, Information Science and Statistics (Springer, New York, 2006)
17Y. LeCun and Y. p. 7 Bishop, Pattern Recognition and Machine Learning, Information Science and Statistics (Springer, New York, 2006). LeCun and Y. LeCun and Y. Bengio, “Convolutional networks for images, speech, and
time series,” in The Handbook of Brain Theory and Neural Networks (MIT
Press, Cambridge, MA, USA, 1998) p.Where the paper says it
Where the paper says it
18L. Jing and Y. p. 7 Bengio, “Convolutional networks for images, speech, and
time series,” in The Handbook of Brain Theory and Neural Networks (MIT
Press, Cambridge, MA, USA, 1998) p. Jing and Y. Jing and Y. Tian, “Self-supervised Visual Feature Learning with Deep
Neural Networks: A Survey,” (2019), arXiv:1902.Where the paper says it
Where the paper says it
19G. Carleo, I Cirac K Cranmer L Daudet M Schuld N Tishby L VogtMaranto and L Zdeborová “Machine learning and the physical sciences” Reviews of Modern Physics 91, 045002 (2019)
20A. Paszke, S Gross F Massa A Lerer J Bradbury G Chanan T Killeen Z Lin N Gimelshein L Antiga A Desmaison A Köpf E Yang Z DeVito M Raison A Tejani S Chilamkurthy B Steiner L Fang J Bai and S Chintala “PyTorch: An Imperative Style, High-Performance Deep Learning Library” (2019), arXiv:1912. p. 7 Bai,
and S. Chintala, “PyTorch: An Imperative Style, High-Performance Deep
Learning Library,” (2019), arXiv:1912.Where the paper says it
21F. Pedregosa G Varoquaux A Gramfort V Michel B Thirion O Grisel M Blondel P Prettenhofer R Weiss V Dubourg J Vanderplas A Passos D Cournapeau M Brucher M Perrot and E Duchesnay “Scikit-learn: Machine learning in Python” Journal of Machine Learning Research 12, 2825–2830 (2011)
26K. Pearson “LIII. p. 7 Yang, “A Survey on Transfer Learning,” IEEE Transactions
on Knowledge and Data Engineering 22, 1345–1359 (2010). Pearson, “LIII.Where the paper says it
27L. McInnes J Healy and J Melville “UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction” (2020), arXiv:1802. p. 7 Healy, and J. Melville, “UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction,” (2020),
arXiv:1802.Where the paper says it
28R. J. G. B. Campello D Moulavi A Zimek and J Sander “Hierarchical Density Estimates for Data Clustering, Visualization, and Outlier Detection” ACM Transactions on Knowledge Discovery from Data 10, 1–51 (2015)
29M Ester H.-P. Kriegel J Sander and X Xu “A density-based algorithm for discovering clusters in large spatial databases with noise” in Proceedings of the Second International Conference on Knowledge Discovery and Data Mining, KDD’96 (AAAI Press, Portland, Oregon, 1996) pp. p. 7 Sander, and X. Xu, “A density-based algorithm
for discovering clusters in large spatial databases with noise,” in Proceedings of the Second International Conference on Knowledge Discovery and
Data Mining, KDD’96 (AAAI Press, Portland, Oregon, 1996) pp.Where the paper says it
30“Fcluster — SciPy v1.16.2 Manual”
31L Breiman “Random Forests” Machine Learning 45, 5–32 (2001)
32L Bottou “Large-Scale Machine Learning with Stochastic Gradient Descent” in Proceedings of COMPSTAT’2010, edited by Y. Lechevallier and G Saporta (Physica-Verlag HD, Heidelberg, 2010) pp. p. 7 Lechevallier and
G. Saporta (Physica-Verlag HD, Heidelberg, 2010) pp.Where the paper says it
33A Dempster F Petitjean and G I Webb “ROCKET: Exceptionally fast and accurate time series classification using random convolutional kernels” Data Mining and Knowledge Discovery 34, 1454–1495 (2020), arXiv:1910.13051 [cs]
34G. E. Hinton and R. p. 7 13051 [cs]. Hinton and R.Where the paper says it
35D. P. Kingma and M Welling “Auto-Encoding Variational Bayes” (2022), arXiv:1312. p. 7 Kingma and M. Welling, “Auto-Encoding Variational Bayes,” (2022),
arXiv:1312.Where the paper says it
36P Vincent H Larochelle I Lajoie Y Bengio and P.-A Manzagol “Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion” Journal of Machine Learning Research 11, 3371–3408 (2010)
37F. Weißer M Baur and W Utschick “Unsupervised Parameter Estimation using Model-based Decoder” (2023), arXiv:2211. p. 7 Baur, and W. Utschick, “Unsupervised Parameter Estimation
using Model-based Decoder,” (2023), arXiv:2211.Where the paper says it
38A Madry A Makelov L Schmidt D Tsipras and A Vladu “Towards Deep Learning Models Resistant to Adversarial Attacks” (2019), arXiv:1706.06083 [stat]
39S Hochreiter and J Schmidhuber “Long Short-Term Memory” Neural Computation 9, 1735–1780 (1997)
40J Duris D Kennedy A Hanuka J Shtalenkova A Edelen P Baxevanis A Egger T Cope M McIntire S Ermon and D Ratner “Bayesian Optimization of a Free-Electron Laser” Physical Review Letters 124, 124801 (2020)
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Create a self-supervised training pipeline that learns representations of raw detector pulses to derive labels without manual labeling, enabling subsequent supervised classification steps. This addresses the challenge of
lack of labeled data
by allowingreal pulses and artifacts follow recognizable patterns that can be discovered via unsupervised learning methods.
-
Implement a multi-stage classification system for pulse rejection using dimensionality reduction (PCA and UMAP) followed by hierarchical density-based clustering (HDBSCAN) to generate labels, which are then used to train a Random Forest Classifier. This provides
robust and scalable solutions for MMC signal processing
by automating the identification of artifacts. -
Develop a feature extraction method using a Convolutional Variational Autoencoder (CVAE) for pulse shape analysis, which learns a
compact representation of the pulse shapes via unsupervised reconstruction.
This latent space regression allows for automated extraction of features like amplitude and trigger time, overcoming the need forhand-optimized
traditional filters. -
Establish a simulation-to-real domain adaptation strategy by training models on simulated data and then fine-tuning them with a small subset of real detector data. This leverages
transfer learning methodology to bridge the gap
between simulation and reality, allowing models to generalize effectively to real detector signals while ensuring performance is maintained. -
Integrate Long Short-Term Memory (LSTM) networks for time series analysis to model and correct for temporal drifts and discontinuities in the detector response. This future work aims to improve gain-drift correction by addressing
temporal drifts and discontinuities in the detector response
during online monitoring. -
Implement a bespoke Bayesian optimization framework to target hardware optimization of readout SQUIDs using real-time feedback from detector signals. This will allow for
automatic tuning of the hardware parameters to achieve optimal performance based on real-time feedback,
moving beyond manual setup procedures.
Abstract
Metallic Magnetic Calorimeters are a promising tool for high-precision X-ray spectroscopy. However the signal processing and calibration pipeline required to exploit their full potential is complex and difficult to scale for large, highly pixelated detector arrays. In this work, we investigate Machine Learning methods for maXs-series detectors, focusing on pulse classification and artifact rejection as well as pulse-shape analysis and feature extraction. By combining unsupervised label auto-discovery with supervised classification and regression, we obtain robust and scalable artifact rejection for high-confidence template building. For ML-based feature extraction, our results constitute a promising proof of principle: in simulation, the ML estimator remains competitive across realistic pulse-shape and noise variation. On real detector data, the method reaches performance comparable to the hand-optimized FIR baseline before gain-drift correction, while further domain adaptation is required to match the established manual analysis pipeline. Both results represent important steps toward a fully automated and scalable data processing procedure.
Sources
- Self-supervised Visual Feature Learning with Deep Neural Networks: A Survey
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- The Mythos of Model Interpretability
- Understanding deep learning requires rethinking generalization
- Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World
- UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
- ROCKET: Exceptionally fast and accurate time series classification using random convolutional kernels
- Auto-Encoding Variational Bayes
- Unsupervised Parameter Estimation using Model-based Decoder
- Towards Deep Learning Models Resistant to Adversarial Attacks
- Deep Clustering for Unsupervised Learning of Visual Features
Related papers
- Cryogenic Mechanical Loss of GaAs/AlGaAs Crystalline Coatings
- The physical design of the superfluid helium-4 UCN source at CSNS
- A new diffuse reflector filament for additive manufacturing of 3D printing finely-segmented plastic scintillator
- Characterization of CRYO ASIC for charge readout in the nEXO experiment
- Reconstructing High-Fidelity Light Yield Maps for Surface LArTPCs Using Crossing Cosmic Muons
- OASIS-ERIS: Open Acquisition System for IEPE Sensors - Electrically Refined and Inter-device Synchronization