madnis 1.0.0

Library for neural importance sampling

Neural Multi-Channel Importance Sampling

MadNIS is a Python library for neural multi-channel importance sampling based on PyTorch. It will be used for Monte Carlo LHC event generation in future versions of MadGraph. This repository provides the MadNIS code as a stand-alone library that can be applied to arbitrary Monte Carlo integration and importance sampling tasks.

This repository contains a refactored version of the code used in our publication The MadNIS reloaded. It is still under active development and will receive frequent updates and bugfixes.

The documentation of the madnis package can be found under docs.madnis.ai.

Installation

You can either install the latest release using pip

pip install madnis

or clone the repository and install the package in dev mode

# clone the repository
git clone https://github.com/MadGraphTeam/madnis.git
# then install in dev mode
cd madnis
pip install --editable .

Citation

If you use this code or parts of it, please cite:

@article{Heimel:2023ngj,
  author = "Heimel, Theo and Huetsch, Nathan and Maltoni, Fabio and Mattelaer, Olivier and Plehn, Tilman and Winterhalder, Ramon",
  title = "{The MadNIS reloaded}",
  eprint = "2311.01548",
  archivePrefix = "arXiv",
  primaryClass = "hep-ph",
  reportNumber = "IRMP-CP3-23-56, MCNET-23-12",
  doi = "10.21468/SciPostPhys.17.1.023",
  journal = "SciPost Phys.",
  volume = "17",
  number = "1",
  pages = "023",
  year = "2024"}