Jet Single Shot Detection

TL;DR

This paper introduces Jet-SSD, a CNN-based object detection method for jet reconstruction at CERN, demonstrating effective quantization with Ternary Weight Networks that maintain high performance.

Contribution

The paper presents a novel application of Single Shot Detection to jet reconstruction and introduces quantization with Ternary Weight Networks for efficient model deployment.

Findings

Jet-SSD effectively localizes and classifies jets in calorimeter images.

Quantized Ternary Weight Networks closely match full-precision performance.

The approach enables efficient and accurate jet detection at CERN.

Abstract

We apply object detection techniques based on Convolutional Neural Networks to jet reconstruction and identification at the CERN Large Hadron Collider. In particular, we focus on CaloJet reconstruction, representing each event as an image composed of calorimeter cells and using a Single Shot Detection network, called Jet-SSD. The model performs simultaneous localization and classification and additional regression tasks to measure jet features. We investigate Ternary Weight Networks with weights constrained to {-1, 0, 1} times a layer- and channel-dependent scaling factors. We show that the quantized version of the network closely matches the performance of its full-precision equivalent.