Multi-scale cross-attention transformer encoder for event classification

2023/12/31 by Hammad, A., Moretti, S., Nojiri, M. · 5 citations
#FOS: Physical sciences #High Energy Physics - Phenomenology (hep-ph)

paper · doi:10.48550/arxiv.2401.00452

Abstract

We deploy an advanced Machine Learning (ML) environment, leveraging a multi-scale cross-attention encoder for event classification, towards the identification of the gg→ H→ hh→ b b b b process at the High Luminosity Large Hadron Collider (HL-LHC), where h is the discovered Standard Model (SM)-like Higgs boson and H a heavier version of it (with m_H>2m_h). In the ensuing boosted Higgs regime, the final state consists of two fat jets. Our multi-modal network can extract information from the jet substructure and the kinematics of the final state particles through self-attention transformer layers. The diverse learned information is subsequently integrated to improve classification performance using an additional transformer encoder with cross-attention heads. We ultimately prove that our approach surpasses in performance current alternative methods used to establish sensitivity to this process, whether solely based on kinematic analysis or else on a combination of this with mainstream ML approaches. Then, we employ various interpretive methods to evaluate the network results, including attention map analysis and visual representation of Gradient-weighted Class Activation Mapping (Grad-CAM). Finally, we note that the proposed network is generic and can be applied to analyse any process carrying information at different scales. Our code is publicly available for generic use.

Cited by

• Interference Effects in Resonant Standard Model di-Higgs Production and Decay into 4b Final States: the Role of Machine Learning Analysis
• Quantifying vacuum-like jets in heavy-ion collisions: a Machine Learning study
• Explaining Data Anomalies over the NMSSM Parameter Space with Deep Learning Techniques
• Deep learning approaches to top FCNC couplings to photons at the LHC
• ALCo-FM: Adaptive Long-Context Foundation Model for Accident Prediction

Related

• Deep Learning to Improve the Sensitivity of Higgs Pair Searches in the 4b Channel at the LHC
• Transformer networks for Heavy flavor jet tagging
• Review of Machine Learning for Real-Time Analysis at the Large Hadron Collider experiments ALICE, ATLAS, CMS and LHCb
• Benchmarking Machine Learning Techniques with Di-Higgs Production at the LHC
• Interpreting Transformers for Jet Tagging