MadNIS -- Neural Multi-Channel Importance Sampling

TL;DR

This paper introduces MadNIS, a neural importance sampling method combining multi-channel weights with normalizing flows, enhancing numerical integration accuracy for complex particle physics processes.

Contribution

It develops a bi-directional invertible network approach that integrates online and buffered training for improved importance sampling in high-energy physics.

Findings

Enhanced integration accuracy for Drell-Yan processes.

Efficient handling of expensive integrands.

Combines machine learning with classical importance sampling methods.

Abstract

Theory predictions for the LHC require precise numerical phase-space integration and generation of unweighted events. We combine machine-learned multi-channel weights with a normalizing flow for importance sampling, to improve classical methods for numerical integration. We develop an efficient bi-directional setup based on an invertible network, combining online and buffered training for potentially expensive integrands. We illustrate our method for the Drell-Yan process with an additional narrow resonance.