Feature Selection with Distance Correlation

Autor:	Das, Ranit, Kasieczka, Gregor, Shih, David
Rok vydání:	2022
Předmět:	High Energy Physics - Phenomenology Computer Science - Machine Learning High Energy Physics - Experiment Physics - Data Analysis Statistics and Probability
Druh dokumentu:	Working Paper
Popis:	Choosing which properties of the data to use as input to multivariate decision algorithms -- a.k.a. feature selection -- is an important step in solving any problem with machine learning. While there is a clear trend towards training sophisticated deep networks on large numbers of relatively unprocessed inputs (so-called automated feature engineering), for many tasks in physics, sets of theoretically well-motivated and well-understood features already exist. Working with such features can bring many benefits, including greater interpretability, reduced training and run time, and enhanced stability and robustness. We develop a new feature selection method based on Distance Correlation (DisCo), and demonstrate its effectiveness on the tasks of boosted top- and $W$-tagging. Using our method to select features from a set of over 7,000 energy flow polynomials, we show that we can match the performance of much deeper architectures, by using only ten features and two orders-of-magnitude fewer model parameters. Comment: 14 pages, 8 figures, 3 tables
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2212.00046 Zobrazit plný text záznamu View this record from Arxiv