Using neural networks for efficient evaluation of high multiplicity scattering amplitudes

Abstract

Precision theoretical predictions for high multiplicity scattering rely on the evaluation of increasingly complicated scattering amplitudes which come with an extremely high CPU cost. For state-of-the-art processes this can cause technical bottlenecks in the production of fully differential distributions. In this article we explore the possibility of using neural networks to approximate multi-variable scattering amplitudes and provide efficient inputs for Monte Carlo integration. We focus on QCD corrections to e + e − → jets up to one-loop and up to five jets. We demonstrate reliable interpolation when a series of networks are trained to amplitudes that have been divided into sectors defined by their infrared singularity structure. Complete simulations for one-loop distributions show speed improvements of at least an order of magnitude over a standard approach.

Per-object systematics using deep-learned calibration

“…However, our quantitative analysis of Bayesian top taggers encountered practical limitations, for instance that the jet energy scale simultaneously affects the central value and the error bar of the probabilistic output. A similar study of uncertainties just appeared for a matrix element regression task [46].…”

Invertible networks or partons to detector and back again

“…Technically, we propose to use invertible networks (INNs) [11][12][13] to invert part of the LHC simulation chain. This application builds on a long list of one-directional applications of generative or similar networks to LHC simulations, including phase space integration [14,15], amplitudes [16,17], event generation [18][19][20][21][22], event subtraction [23], detector simulations [24][25][26][27][28][29][30][31][32], parton showers [33][34][35][36], or searches for physics beyond the Standard Model [37]. INNs are an alternative class of generative networks, based on normalizing flows [38][39][40][41].…”

Neural resampler for Monte Carlo reweighting with preserved uncertainties

“…These approaches, however, require learning the full phase space density, instead of just the likelihood ratio, which is significantly more complicated than the neural resampling approach presented here. It is also worth mentioning that neural networks and other machine learning techniques have been studied to improve other components of event generation, including parton density modeling [65,66], phase space generation [67][68][69][70][71], matrix element calculations [72,73], and more [74][75][76].…”