Toward Machine Learning Optimization of Experimental Design - Archive ouverte HAL
Article Dans Une Revue Nucl.Phys.News Année : 2021

Toward Machine Learning Optimization of Experimental Design

Atilim Güneş Baydin
  • Fonction : Auteur
Kyle Cranmer
  • Fonction : Auteur
Pablo de Castro Manzano
  • Fonction : Auteur
Christophe Delaere
  • Fonction : Auteur
Denis Derkach
  • Fonction : Auteur
Tommaso Dorigo
  • Fonction : Auteur
Andrea Giammanco
  • Fonction : Auteur
Jan Kieseler
  • Fonction : Auteur
Lukas Layer
  • Fonction : Auteur
Gilles Louppe
  • Fonction : Auteur
Fedor Ratnikov
  • Fonction : Auteur
Giles Strong
  • Fonction : Auteur
Mia Tosi
  • Fonction : Auteur
Andrey Ustyuzhanin
  • Fonction : Auteur
Pietro Vischia
  • Fonction : Auteur
Hevjin Yarar
  • Fonction : Auteur

Résumé

The effective design of instruments that rely on the interaction of radiation with matter for their operation is a complex task. A full optimization of the many parameters involved may still be sought by leveraging recent progress in computer science. Key to such a goal is the definition of a utility function that models the true goals of the instrument. Such a function must account for the interplay between physical processes that are intrinsically stochastic in nature and the vast space of possible choices for the physical characteristics of the instrument. The construction of a differentiable model of all the ingredients of the information-extraction procedures, including data collection, detector response, pattern recognition, and all existing constraints, then allows the automatic exploration of the vast space of design choices and the search for their best combination. In this document we succinctly describe the research program of the MODE Collaboration (an acronym for Machine-learning Optimized Design of Experiments), which aims at developing tools based on deep learning techniques to achieve end-to-end optimization of the design of instruments via a fully differentiable pipeline capable of exploring the Pareto-optimal frontier of the utility function. The goal of MODE is to demonstrate those techniques on small-scale applications such as muon tomography or hadron therapy, to then gradually adapt them to the more ambitious task of exploring innovative solutions to the design of detectors for future particle collider experiments.

Dates et versions

hal-03178529 , version 1 (23-03-2021)

Identifiants

Citer

Atilim Güneş Baydin, Kyle Cranmer, Pablo de Castro Manzano, Christophe Delaere, Denis Derkach, et al.. Toward Machine Learning Optimization of Experimental Design. Nucl.Phys.News, 2021, 31 (1), pp.25-28. ⟨10.1080/10619127.2021.1881364⟩. ⟨hal-03178529⟩

Collections

PRES_CLERMONT
221 Consultations
0 Téléchargements

Altmetric

Partager

More