Quark/Gluon Discrimination and Top Tagging with Dual Attention Transformer

Jet tagging is a crucial classification task in high energy physics. Recently the performance of jet tagging has been significantly improved by the application of deep learning techniques. In this work, we propose Particle Dual Attention Transformer for jet tagging, a new transformer architecture which captures both global information and local information simultaneously. Based on the point cloud representation, we introduce the Channel Attention module to the point cloud transformer and incorporates both the pairwise particle interactions and the pairwise jet feature interactions in the attention mechanism. We demonstrate the effectiveness of the P-DAT architecture in classic top tagging and quark-gluon discrimination tasks, achieving competitive performance compared to other benchmark strategies.
Comment: 15 pages, 4 figures, 3 tables