# Lorentz Boost Networks: Autonomous Physics-Inspired Feature Engineering

## TL;DR

This paper introduces Lorentz Boost Networks, a physics-inspired neural network architecture that autonomously constructs meaningful particle combinations and features from collision event data, improving classification performance.

## Contribution

The paper presents a novel two-stage neural network architecture that automatically generates physically meaningful features from particle momenta for event classification.

## Key findings

- LBN achieves leading separation power in classifying ttH and ttbb events.
- LBN autonomously creates physically interpretable particle combinations.
- LBN outperforms domain-unspecific DNNs despite less input information.

## Abstract

We present a two-stage neural network architecture that enables a fully autonomous and comprehensive characterization of collision events by exclusively exploiting the four momenta of final-state particles. We refer to the first stage of the architecture as Lorentz Boost Network (LBN). The LBN allows the creation of particle combinations representing rest frames. The LBN also enables the formation of further composite particles, which are then transformed into said rest frames by Lorentz transformation. The properties of the composite, transformed particles are compiled in the form of characteristic variables that serve as input for a subsequent network. This second network has to be configured for a specific analysis task such as the separation of signal and background events. Using the example of the classification of ttH and ttbb events, we compare the separation power of the LBN approach with that of domain-unspecific deep neural networks (DNN). We observe leading performance with the LBN, even though we provide the DNNs with extensive additional input information beyond the particle four momenta. Furthermore, we demonstrate that the LBN forms physically meaningful particle combinations and autonomously generates suitable characteristic variables.