Design of detectors at the electron ion collider with artificial intelligence

Abstract

Artificial Intelligence (AI) for design is a relatively new but active area of research across many disciplines. Surprisingly when it comes to designing detectors with AI this is an area at its infancy. The electron ion collider is the ultimate machine to study the strong force. The EIC is a large-scale experiment with an integrated detector that extends for about ±35 meters to include the central, far-forward, and far-backward regions. The design of the central detector is made by multiple sub-detectors, each in principle characterized by a multidimensional design space and multiple design criteria also called objectives. Simulations with Geant4 are typically compute intensive, and the optimization of the detector design may include non-differentiable terms as well as noisy objectives. In this context, AI can offer state of the art solutions to solve complex combinatorial problems in an efficient way. In particular, one of the proto-collaborations, ECCE, has explored during the detector proposal the possibility of using multi-objective optimization to design the tracking system of the EIC detector. This document provides an overview of these techniques and recent progress made during the EIC detector proposal. Future high energy nuclear physics experiments can leverage AI-based strategies to design more efficient detectors by optimizing their performance driven by physics criteria and minimizing costs for their realization.

The optimal use of segmentation for sampling calorimeters

“…This research is part of a larger movement to automate detector design with machine learning. Recent works have considered a number of different instruments with a variety of approaches [29][30][31][32][33][34]. The studies presented here could be combined with a point cloud generative model for an end-to-end optimization [35][36][37][38][39][40][41].…”

Artificial Intelligence for imaging Cherenkov detectors at the EIC

“…In [19] studies have been performed utilizing samples produced with FastDIRC for the GlueX DIRC design. In [23] it has been discussed: (a) the possibility of using for training high purity samples directly from real data using specific topologies with, e.g., π, K and p; (b) a potential procedure for data augmentation at any given bin of the particle kinematics, consisting of sampling the expected hit pattern according to the expected photon yield distribution. The main features of DeepRICH can be summarized in the following points: (i) it is fast and provides accurate reconstruction; (ii) it can be extended to multiple particle types (multi-class identification); (iii) it can be generalized to fast simulation, using VAE as a generative model; (iv) it can utilize (x, y, t) patterns if time is measured; can deal with different topologies and different detectors; (v) it deeply learns the detector response (high-purity samples of real data can be injected during the training process).…”

Geometry Optimization for Long-lived Particle Detectors

“…This algorithmic approach may be contrasted with other detector optimization techniques that use an automated optimizer, either combinatorial or gradient-based as appropriate; see e.g. Refs [46][47][48].…”