SMARTHEP: training PhD students in real-time analysis at the LHC and in industry
Johannes Albrecht, Laura Boggia, Leon Bozianu, Andrew Carey, Carlos Cocha, Caterina Doglioni, James Andrew Gooding, Joachim Hansen, Patin Inkaew, Kaare Iversen, Pratik Jawahar, Henning Kirschenmann, Daniel Magdalinski, Alice Ohlson, Micol Olocco, Henrique Piñeiro Monteagudo, Steven Schramm, Mike Sokoloff, Alexandros Sopasakis, Leonardo Taccari, Sten Åstrand
TL;DR
The paper addresses the challenge of unprecedented data volumes at the LHC and in industry demanding real-time, data-efficient analysis. It presents SMARTHEP, an MSCA ITN program that trained 12 ESRs across Europe with dual supervision and industry secondments to develop Real-Time Analysis techniques applicable to LHC triggers and commercial applications, including edge computing. The network delivered technical outcomes such as FPGA-vs-GPU benchmarking for track reconstruction, Graph Neural Networks for tracking, anomaly-detection pipelines, and two whitepapers on triggers and ML in RTA, plus substantial industry impact through autonomous driving and fraud detection applications. This cross-sector, training-centric approach demonstrates tangible benefits for both high-energy physics experiments and industry, with enduring collaborations and a framework that can persist beyond the current funding cycle via associated coalitions and COST actions.
Abstract
In this invited Editorial for Software and Computing for Big Science, we describe the SMARTHEP Innovative Training Network funded via the Marie Skłodowska-Curie Actions between 2021 and 2025. SMARTHEP trained 12 PhD students to advance machine learning and real-time analysis in high-energy physics experiments and industrial applications. We present the perspective of students, supervisors, and external observers of the network, concerning the work done within the network, the added value compared to ``typical'' PhD positions, and the emerging themes and directions from our experiences in the past four years.