En poursuivant votre navigation sur ce site, vous acceptez le dépôt de cookies dans votre navigateur. (En savoir plus)

PhD student in data, knowledge, learning and interactions (M / F)

This offer is available in the following languages:
Français - Anglais

Date Limite Candidature : jeudi 5 novembre 2020

Assurez-vous que votre profil candidat soit correctement renseigné avant de postuler. Les informations de votre profil complètent celles associées à chaque candidature. Afin d’augmenter votre visibilité sur notre Portail Emploi et ainsi permettre aux recruteurs de consulter votre profil candidat, vous avez la possibilité de déposer votre CV dans notre CVThèque en un clic !

Faites connaître cette offre !

General information

Reference : UMR9012-SYLPRA0-022
Workplace : ORSAY,ORSAY
Date of publication : Thursday, October 15, 2020
Scientific Responsible name : Haÿg GULER
Type of Contract : PhD Student contract / Thesis offer
Contract Period : 36 months
Start date of the thesis : 1 December 2020
Proportion of work : Full time
Remuneration : 2 135,00 € gross monthly

Description of the thesis topic

The performances required for an electron beam for an accelerator are more and more demanding. Performance for free electron laser sources, plasma laser acceleration, or Compton sources is based on the brightness and peak current of the electron beam. To achieve this, it is necessary to fine tune the accelerator in a non-linear fashion: magnetic alignment, “golden-orbit”, RF phase matching, etc. This requires many parameters to optimize the properties of the beam specific to each application.
The objective of this thesis, co-directed by two laboratories (IJCLAB and the Computer Science Research Laboratory, LRI) is to explore and implement automatic learning methods in the context of particle accelerators.
The accelerators concerned are THOMX, whose commissioning will begin shortly, as well as the PHIL / LASERIX platform (PALLAS project).
The thesis topic presented addresses this problem in the context of machine learning, responding to the following challenges:
• Modeling of a stochastic phenomenon
• Active learning (identification of the most informative destructive measurements to perform the dynamic control)
• Reinforcement learning (use of simulators making it possible to predict the behavior of the process in the short term, optimization of the compromise cost of calculating the simulator / precision of predictions).

Bibliographical references:
A. L. Edelen, S. G. Biedron, B. E. Chase, D. Edstrom, S. V. Milton, and
P. Stabile, “Neural Networks for Modeling and Control of Particle Accelerators,” IEEE Trans. Nucl. Sci. 63, 878–897 (2016), arXiv: 1610.06151 [physics.acc-ph].
T. Higo, H. Shoaee, and J Spencer, “Some applications of ai to the problems of accelerator physics,” in Conf. Proc., Vol. 870316 (1986) p. 701.
G. Valentino, R. Bruce, S. Redaelli, R. Rossi, P. Theodoropoulos, and S. Jaster-Merz, “Anomaly Detection for Beam Loss Maps in the Large Hadron Collider,” Proceedings, 8th International Particle Accelerator Conference (IPAC 2017): Copenhagen, Denmark, May 14- 19, 2017, J. Phys. Conf. Ser. 874, 012002 (2017), [, MOPAB010 (2017)].

Work Context

The Irène Joliot-Curie Physics of the 2 Infinities laboratory is a physics laboratory of the two infinities under the supervision of the CNRS, the University of Paris-Saclay and the University of Paris, born in 2020 from the merger of the five UMRs located on the Orsay university campus: the Center for Nuclear and Material Sciences (CSNSM), the Imaging and Modeling Laboratory in Neurobiology and Cancer (IMNC), the Orsay Institute of Nuclear Physics (IPNO), the Linear Accelerator Laboratory (LAL) and the Theoretical Physics Laboratory (LPT).
The research topics of the laboratory are nuclear physics, high energy physics, astroparticles and cosmology, theoretical physics, particle accelerators and detectors as well as research and technical developments and associated applications for energy, health and the environment.
The structure has very important technical capacities (approximately 280 engineers and technicians) in all the major fields required to design, develop and implement the experimental devices necessary for its scientific activity: mechanics, electronics, data processing, instrumentation, techniques of acceleration and techniques of biology. These technical strengths represent a major asset for the design, development and use of the necessary instruments (accelerators and detectors). The presence of research infrastructures and technological platforms assembled on the laboratory site is also a major asset. Finally, around 90 ITAs from the administrative and support services work alongside scientists and engineers.

Constraints and risks

The thesis will be attached to the STIC doctoral school (Information and Communication Sciences and Technologies)
The thesis work will take place in the two laboratories: The Irène Joliot-Curie Physics Laboratory of the 2 Infinites as well as the Computer Science Research Laboratory.

Additional Information

Expected skills:
• Good knowledge of programming (Python, ...)
• Good understanding of object oriented programming, advanced algorithms, data structures, etc.
• Very good knowledge of machine learning (Machine learning, deep learning, Reinforcement learning)
• Comfortable with data visualization tools like Scipy, Matlab, etc.
• General physics
• The physics of accelerators is a plus
• Level of English required: Upper intermediate

We talk about it on Twitter!