Fast Inclusive Flavour Tagging at LHCb

TL;DR

This paper introduces a new DeepSet-based classifier for B meson flavor tagging at LHCb, significantly improving training and inference speed while maintaining or enhancing tagging performance, especially suitable for real-time trigger environments.

Contribution

The work presents a novel DeepSet architecture for inclusive flavor tagging, achieving faster training and inference, and demonstrating its effectiveness on upgraded LHCb detector simulations.

Findings

Training time reduced from hours to 10 minutes.

Inference speed increased by a factor of 4-5.

Comparable or improved tagging performance on upgraded detector simulations.

Abstract

The task of identifying B meson flavor at the primary interaction point in the LHCb detector is crucial for measurements of mixing and time-dependent CP violation. Flavour tagging is usually done with a small number of expert systems that find important tracks to infer the B meson flavour from. Recent advances show that replacing all of those expert systems with one ML algorithm that considers all tracks in an event yields an increase in tagging power. However, training the current classifier takes a long time and is not suitable for use in real-time triggers. In this work we present a new classifier, based on the DeepSet architecture. With the right inductive bias of permutation invariance, we achieve great speedups in training (multiple hours vs 10 minutes), a factor of 4-5 speed-up in inference for use in real time environments like the trigger and less tagging asymmetry. For the first time we investigate and compare performances of these Inclusive Flavor Taggers on simulation of the upgraded LHCb detector for the third run of the LHC.