2007/04/30 by J. Bastos, João A. Bastos, Y. Liu +1
Chemistry · Mathematics · Physics and Astronomy · #Alternating decision tree #Artificial intelligence #Artificial neural network #Cascade #Chemistry #Chromatography #Computer science #Decision tree #Decision tree learning #High-Energy Particle Collisions Research #Machine learning #Mathematics #Monte Carlo method #Particle Detector Development and Performance #Particle physics theoretical and experimental studies #Statistics #physics.data-an
paper · pdf · doi:10.1088/1748-0221/2/11/p11007
published as JINST 2:P11007,2007 · 14 pages, 12 figures, revised version
openalex publication_date 2007/11/27 · arxiv created 2007/11/28 · arxiv updated 2011/01/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
This paper compares the performance of artificial neural networks and boosted decision trees, with and without cascade training, for tagging b-jets in a collider experiment. It is shown, using a Monte Carlo simulation of WH → l ν q events, that for a b-tagging efficiency of 50%, the light jet rejection power given by boosted decision trees without cascade training is about 55% higher than that given by artificial neural networks. The cascade training technique can improve the performance of boosted decision trees and artificial neural networks at this b-tagging efficiency level by about 35% and 80% respectively. We conclude that the cascade trained boosted decision trees method is the most promising technique for tagging heavy flavours at collider experiments.