TMVA - Toolkit for Multivariate Data Analysis

TL;DR

TMVA is a comprehensive toolkit integrated with ROOT that offers various multivariate classification and regression algorithms, enabling efficient data analysis in high-energy physics with flexible data handling and improved performance features.

Contribution

It introduces TMVA 4, extending capabilities to multivariate regression and flexible data handling, with new combined methods and a generalized boosting approach.

Findings

Enhanced multivariate regression capabilities

Flexible data handling for combined methods

Introduction of a generalized boosting method

Abstract

In high-energy physics, with the search for ever smaller signals in ever larger data sets, it has become essential to extract a maximum of the available information from the data. Multivariate classification methods based on machine learning techniques have become a fundamental ingredient to most analyses. Also the multivariate classifiers themselves have significantly evolved in recent years. Statisticians have found new ways to tune and to combine classifiers to further gain in performance. Integrated into the analysis framework ROOT, TMVA is a toolkit which hosts a large variety of multivariate classification algorithms. Training, testing, performance evaluation and application of all available classifiers is carried out simultaneously via user-friendly interfaces. With version 4, TMVA has been extended to multivariate regression of a real-valued target vector. Regression is invoked through the same user interfaces as classification. TMVA 4 also features more flexible data handling allowing one to arbitrarily form combined MVA methods. A generalised boosting method is the first realisation benefiting from the new framework.