# CapsNets Continuing the Convolutional Quest

## TL;DR

This paper demonstrates the effectiveness of capsule networks in high-energy physics for resonance detection, background discrimination, interpretability, and integrating multiple data sources beyond calorimeter images.

## Contribution

It introduces capsule networks for particle physics analysis, showing their advantages over traditional convolutional networks in resonance identification and interpretability.

## Key findings

- Capsule networks outperform standard CNNs in resonance detection.
- Multi-class capsules effectively distinguish signal from background.
- Capsule networks can integrate diverse data types, such as overlaying images.

## Abstract

Capsule networks are ideal tools to combine event-level and subjet information at the LHC. After benchmarking our capsule network against standard convolutional networks, we show how multi-class capsules extract a resonance decaying to top quarks from both, QCD di-jet and the top continuum backgrounds. We then show how its results can be easily interpreted. Finally, we use associated top-Higgs production to demonstrate that capsule networks can work on overlaying images to go beyond calorimeter information.