Jet Single Shot Detection

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 CaloJet reconstruction, representing each event as an image composed of calorimeter cells using a Single Shot Detection network, called Jet-SSD. The model performs simultaneous localization classification additional regression tasks measure features. investigate TernaryWeight with weights constrained {-1, 0, 1} times layer- channel-dependent scaling factors. show that quantized version network closely matches performance its full-precision equivalent.