Learning to increase matching efficiency in identifying additional b-jets in the t(t)over-barb(b)over-bar

Abstract

The t (t) over barH(b (b) over bar) process is an essential channel in revealing the Higgs boson properties; however, its final state has an irreducible background from the t (t) over barb (b) over bar process, which produces a top quark pair in association with a b quark pair. Therefore, understanding the t (t) over barb (b) over bar process is crucial for improving the sensitivity of a search for the t (t) over barH(b (b) over tilde) process. To this end, when measuring the differential cross section of the t (t) over barb (b) over bar process, we need to distinguish the b-jets originating from top quark decays and additional b-jets originating from gluon splitting. In this paper, we train deep neural networks that identify the additional b-jets in the t (t) over barb (b) over bar events under the supervision of a simulated t (t) over barb (b) over bar event data set in which true additional b-jets are indicated. By exploiting the special structure of the t (t) over barb (b) over bar event data, several loss functions are proposed and minimized to directly increase matching efficiency, i.e., the accuracy of identifying additional b-jets. We show that, via a proof-of-concept experiment using synthetic data, our method can be more advantageous for improving matching efficiency than the deep learning-based binary classification approach presented in [1]. Based on simulated t (t) over barb (b) over bar event data in the lepton+jets channel from pp collision at root s = 13 TeV, we then verify that our method can identify additional b-jets more accurately: compared with the approach in [1], the matching efficiency improves from 62.1% to 64.5% and from 59.9% to 61.7% for the leading order and the next-to-leading order simulations, respectively.

The t¯tH(bb) process is an essential channel in revealing the Higgs boson properties; however, its final state has an irreducible background from the t¯tbb process, which produces a top quark pair in association with a b quark pair. Therefore, understanding the t¯tbb process is crucial for improving the sensitivity of a search for the t ¯ ¯tH(bb) process. To this end, when measuring the differential cross section of the t¯tbb process, we need to distinguish the b-jets originating from top quark decays and additional b-jets originating from gluon splitting. In this paper, we train deep neural networks that identify the additional b-jets in the t¯tbb events under the supervision of a simulated t¯tbb event data set in which true additional b-jets are indicated. By exploiting the special structure of the t¯tbb event data, several loss functions are proposed and minimized to directly increase matching efficiency, i.e., the accuracy of identifying additional b-jets. We show that, via a proof-of-concept experiment using synthetic data, our method can be more advantageous for improving matching efficiency than the deep learning-based binary classification approach presented in [1]. Based on simulated t¯tbb event data in the lepton+jets channel from pp collision at ¯ √s 13 TeV, we then verify that our method can identify additional b-jets more accurately: compared with the approach in [1], the matching efficiency improves from 62.1% to 64.5% and from 59.9% to 61.7% for the leading order and the next-to-leading order simulations, respectively.