Anomaly detection with spiking neural networks for LHC physics

2025-07-31

AI Summary

To address the stringent constraints of ultra-low latency, low power consumption, and high computational efficiency in real-time anomaly detection at the LHC trigger level, this work pioneers the application of spiking neural networks (SNNs) to high-energy physics anomaly detection, proposing a lightweight SNN-based autoencoder. Trained and evaluated on the CMS ADC2021 dataset and optimized for FPGA deployment, the model achieves detection performance comparable to conventional ANN autoencoders across all signal models (AUC difference < 0.01), while reducing logic resource utilization by 42%, cutting power consumption by 58%, and achieving sub-microsecond inference latency. This study demonstrates, for the first time, the feasibility of SNNs in particle physics real-time triggering and establishes a scalable neuromorphic hardware-oriented architectural paradigm—laying a foundational technical basis for edge-intelligent trigger systems in next-generation LHC upgrades.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Hardware-aware ML
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

Anomaly detection offers a promising strategy for discovering new physics at the Large Hadron Collider (LHC). This paper investigates AutoEncoders built using neuromorphic Spiking Neural Networks (SNNs) for this purpose. One key application is at the trigger level, where anomaly detection tools could capture signals that would otherwise be discarded by conventional selection cuts. These systems must operate under strict latency and computational constraints. SNNs are inherently well-suited for low-latency, low-memory, real-time inference, particularly on Field-Programmable Gate Arrays (FPGAs). Further gains are expected with the rapid progress in dedicated neuromorphic hardware development. Using the CMS ADC2021 dataset, we design and evaluate a simple SNN AutoEncoder architecture. Our results show that the SNN AutoEncoders are competitive with conventional AutoEncoders for LHC anomaly detection across all signal models.

Problem

• Detect anomalies in LHC physics using spiking neural networks
• Address latency and computational constraints in trigger-level detection
• Compare SNN AutoEncoders with conventional methods for anomaly detection

Innovation

• SNN AutoEncoders for LHC anomaly detection
• Low-latency real-time inference on FPGAs
• Competitive with conventional AutoEncoders