Scaling laws for amplitude surrogates

2026-01-19

AI Summary

This study systematically investigates the scaling laws of neural networks in the context of surrogate modeling for particle physics scattering amplitudes, establishing for the first time a clear relationship between these laws and the number of external particles. Through large-scale computational experiments, the authors analyze how model performance varies with training data volume, computational resources, and model size. The results confirm the universality of scaling laws in this domain and demonstrate their effectiveness in guiding the construction of high-precision amplitude surrogate models. This work provides both theoretical grounding and practical pathways for future high-energy physics simulations leveraging machine learning.

Technology Category

• Machine Learning: Scalability of ML Systems
• Natural Language Processing: Learning & Optimization for NLP
• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

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.

Problem

• scaling laws
• amplitude surrogates
• neural network performance
• particle physics
• external particles

Innovation

• scaling laws
• amplitude surrogates
• neural networks
• particle physics
• external particles