Machine learning for surface prediction in ACTS

TL;DR

This paper explores machine learning techniques to improve surface prediction for detector navigation in track reconstruction within the ACTS tracking toolkit, comparing various neural network training approaches.

Contribution

It introduces neural network-based methods for surface prediction and evaluates their performance in the context of ACTS, advancing ML integration in detector navigation.

Findings

Neural networks can effectively predict detector surfaces.

Different training approaches yield varying accuracy levels.

The study demonstrates potential for ML to enhance track reconstruction.

Abstract

We present an ongoing R&D activity for machine-learning-assisted navigation through detectors to be used for track reconstruction. We investigate different approaches of training neural networks for surface prediction and compare their results. This work is carried out in the context of the ACTS tracking toolkit.