New directions for surrogate models and differentiable programming for High Energy Physics detector simulation

TL;DR

This paper explores innovative surrogate models and differentiable programming techniques to address the high computational costs of high energy physics detector simulations, aiming to improve efficiency and scalability.

Contribution

It introduces new approaches and ongoing efforts in applying machine learning-based surrogate models and differentiable programming to particle detector simulation.

Findings

Potential reduction in simulation computational costs

Enhanced control and scalability in simulation routines

Integration of machine learning with physics simulations

Abstract

The computational cost for high energy physics detector simulation in future experimental facilities is going to exceed the current available resources. To overcome this challenge, new ideas on surrogate models using machine learning methods are being explored to replace computationally expensive components. Additionally, differentiable programming has been proposed as a complementary approach, providing controllable and scalable simulation routines. In this document, new and ongoing efforts for surrogate models and differential programming applied to detector simulation are discussed in the context of the 2021 Particle Physics Community Planning Exercise (`Snowmass').