We present a novel approach to this class of problem, based on neural networks using a generalized attention mechanism, that we call Symmetry Preserving Attention Networks (SPA-Net) to outperform existing state-of-the-art methods for jet-parton assignment.
This Simulation-level fit based on Reweighting Generator-level events with Neural networks (SRGN) is demonstrated using simulated datasets for a variety of examples including a simple Gaussian random variable, parton shower tuning, and the top quark mass extraction.
Combining jet substructure and event information with modern machine learning, we demonstrate the ability to focus on particular production modes.
We studied the application of deep networks to heavy-flavor jets classification byadding lower-level track and vertex information to the high-level features and showed a significant boost in performance compared to the state of the art.