Lightweight Jet Reconstruction and Identification as an Object Detection Task

TL;DR

This paper introduces PFJet-SSD, a deep learning object detection approach for jet reconstruction at the LHC, achieving faster and more accurate results than traditional methods, with optimized quantization for real-time processing.

Contribution

It presents a novel end-to-end deep learning method for jet detection, combining localization, classification, and regression in a single network optimized for real-time use.

Findings

Ternary quantization nearly matches full-precision accuracy.

PFJet-SSD outperforms traditional rule-based algorithms.

Achieves low inference latency suitable for real-time applications.

Abstract

We apply object detection techniques based on deep convolutional blocks to end-to-end jet identification and reconstruction tasks encountered at the CERN Large Hadron Collider (LHC). Collision events produced at the LHC and represented as an image composed of calorimeter and tracker cells are given as an input to a Single Shot Detection network. The algorithm, named PFJet-SSD performs simultaneous localization, classification and regression tasks to cluster jets and reconstruct their features. This all-in-one single feed-forward pass gives advantages in terms of execution time and an improved accuracy w.r.t. traditional rule-based methods. A further gain is obtained from network slimming, homogeneous quantization, and optimized runtime for meeting memory and latency constraints of a typical real-time processing environment. We experiment with 8-bit and ternary quantization, benchmarking their accuracy and inference latency against a single-precision floating-point. We show that the ternary network closely matches the performance of its full-precision equivalent and outperforms the state-of-the-art rule-based algorithm. Finally, we report the inference latency on different hardware platforms and discuss future applications.