用于CMS一级触发器中实时异常检测的正则化流知识蒸馏
Knowledge Distillation of a Normalising Flow for Real-Time Anomaly Detection at LHC Level-1 Trigger
浏览论文内容
中文总结 AI 辅助
本研究提出一种将正则化流教师模型的知识蒸馏至紧凑神经网络的方法，用于CMS一级触发器的实时异常检测，实现高精度、低延迟的FPGA部署，支持全速率新物理搜索。
中文摘要 AI 辅助
我们提出了一种在LHC事件处理的硬件第一阶段（一级触发器）中进行无监督异常检测的新策略。一个仅在标准模型（SM）事件上训练的正则化流，基于其精确的负对数似然提供教师异常分数，该分数被蒸馏到一个适合FPGA部署的紧凑神经网络中。教师模型在基准数据集上达到了最先进的性能，采用了基于p值的严格异常分数定义。一个简单的两层隐藏层学生模型在四个超越标准模型的基准测试中复现了教师模型的性能，AUC差异在0.1个百分点以内，同时实现了约325倍的压缩因子。使用PQuantML进行量化感知训练，将模型降至8位精度，AUC变化低于0.06个百分点。使用Alkaid编译为Verilog固件，生成了无DSP和BRAM的FPGA设计，延迟为27-52纳秒。这一模块化流程将异常检测模型的表达能力与实时硬件约束解耦，使得在40 MHz的完整LHC碰撞率下，能够使用复杂架构进行与模型无关的新物理搜索。
英文摘要
We present a new strategy for unsupervised anomaly detection at the hardware-based first stage of event processing (Level-1 trigger) of the Large Hadron Collider (LHC). A normalising flow trained exclusively on Standard Model (SM) events provides a teacher anomaly score based on its exact negative log-likelihood, which is distilled into a compact neural network suitable for field-programmable gate array (FPGA) deployment. The teacher reaches state-of-the-art performance on a benchmark dataset, using a rigorous p-value-based definition of the anomaly score. A simple two-hidden-layer student reproduces the teacher performance across four beyond-the-SM benchmarks to within 0.1 percentage points in area under the receiver operating characteristic curve (AUC), while achieving a compression factor of approximately 325 times. Quantisation-aware training with PQuantML reduces the model to 8-bit precision with AUC changes below 0.06 percentage points. Compilation to Verilog firmware with Alkaid yields FPGA designs that require neither digital signal processors nor block random-access memory and achieve latencies of 27-52 ns. This modular pipeline decouples the expressiveness of the anomaly-detection model from real-time hardware constraints, enabling complex architectures to be used for model-agnostic searches for new physics at the full LHC collision rate of 40 MHz.
发表机构
- Physik-Institut, Universität Zürich (UZH)（苏黎世大学物理研究所）
- European Organization for Nuclear Research (CERN)（欧洲核子研究中心）
- California Institute of Technology（加州理工学院）
- Scuola Normale Superiore（比萨高等师范学院）
- Istituto Nazionale di Fisica Nucleare (INFN), Sezione di Pisa（意大利国家核物理研究所比萨分部）
机构由 AI 辅助整理，请以论文原文为准。
