CUDACPP gives MadGraph data-parallel helicity amplitudes, delivering linear SIMD CPU speed-ups and up to order-of-magnitude GPU speed-ups for high-multiplicity QCD event generation.
Automating scattering amplitudes with chirality flow
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Recently we introduced the chirality-flow formalism, a method which builds on the spinor-helicity formalism and is inspired by the color-flow idea in QCD. With this formalism, Feynman rules and diagrams are simplified to the extent that it is often possible to immediately, by hand, write down a helicity amplitude given a Feynman diagram. In this paper we show that the method can also speed up numerical evaluation of scattering amplitudes by considering $e^+ e^-$ going to $n$ photons in a MadGraph-based tree-level implementation. We find that the computation time is reduced by roughly a factor ten for six photons, and that it scales better with the number of external particles than the default MadGraph5_aMC@NLO implementation. This performance gain is in part attributed to the more compact Lorentz structures involved, and in part due to a transparent choice of gauge reference vectors which reduces the number of Feynman diagrams considered.
citation-role summary
citation-polarity summary
fields
hep-ph 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Data-parallel leading-order event generation in MadGraph5_aMC@NLO
CUDACPP gives MadGraph data-parallel helicity amplitudes, delivering linear SIMD CPU speed-ups and up to order-of-magnitude GPU speed-ups for high-multiplicity QCD event generation.