Scaling laws for amplitude surrogates

TL;DR

This paper explores how scaling laws relate to amplitude surrogates in particle physics, revealing connections to external particles and demonstrating their utility in achieving precision goals.

Contribution

It systematically investigates scaling laws for amplitude surrogates in particle physics, linking scaling coefficients to the number of external particles and showcasing their practical use.

Findings

Scaling coefficients relate to the number of external particles.

Scaling laws help achieve targeted precision.

Demonstrates the utility of scaling laws in particle physics.

Abstract

Scaling laws describing the dependence of neural network performance on the amount of training data, the spent compute, and the network size have emerged across a huge variety of machine learning task and datasets. In this work, we systematically investigate these scaling laws in the context of amplitude surrogates for particle physics. We show that the scaling coefficients are connected to the number of external particles of the process. Our results demonstrate that scaling laws are a useful tool to achieve desired precision targets.