HEP ML Lab: An end-to-end framework for applying machine learning into phenomenology studies

TL;DR

HEP ML LAB, a Python-based, end-to-end framework for phenomenology studies, provides the complete workflow from event generation to performance evaluation, and provides a consistent style of use for different approaches, and proposes an observable naming convention to streamline the data extraction and conversion processes.

Abstract

Recent years have seen the development and growth of machine learning in high energy physics. There will be more effort to continue exploring its full potential. To make it easier for researchers to apply existing algorithms and neural networks and to advance the reproducibility of the analysis, we develop the HEP ML Lab (hml), a Python-based, end-to-end framework for phenomenology studies. It covers the complete workflow from event generation to performance evaluation, and provides a consistent style of use for different approaches. We propose an observable naming convention to streamline the data extraction and conversion processes. In the Keras style, we provide the traditional cut-and-count and boosted decision trees together with neural networks. We take the $W^+$ tagging as an example and evaluate all built-in approaches with the metrics of significance and background rejection. With its modular design, HEP ML Lab is easy to extend and customize, and can be used as a tool for both beginners and experienced researchers.

Figures (14)

• Figure 1: All modules in the hml framework and main classes in each module.
• Figure 2: Initialize Madgraph5.
• Figure 3: Methods of Madgraph5 to generate processes.
• Figure 4: Use launch method and set up all possible parameters for generating events.
• Figure 5: The output of summary method.