LEARNING TO DISCOVER

Organizers
  •  Cécile Germain (LRI, Orsay, France)
  • Vava Gligorov (LPNHE, Paris, France)
  • Eilam Gross (Weizmann, Rehovot, Israël)
  • Michael Kagan (SLAC, USA)
  • Danilo Rezende (Google Deepmin, London, UK)
  • David Rousseau (IJCLab, Orsay, France)
  • Andreas Salzburger (CERN, Switzerland)
  • Jean-Roch Vlimant (CalTech, USA)
  • Slava Voloshynovskyii (U Genève, Switzerland)
Date of start
Date of end
Learning
INDICO WEBSITE

How to register : please read

Welcome to the website of the program. On this page you will find more information on the program, registered participants and other details. In order to apply/register to this program, Institut Pascal first invites you to create an IPa-account, which will serve as a way for you to have direct access to your program registration and allows you to update your preferences directly into the IPa-system. Making an account is a two-step procedure : First please click here : [registration]. After you have created your profile, you will receive an email containing a confirmation link. After clicking on this link, you will be prompted to enter your password. Once your IPa-account is created, you are able to login and sign up for the program of your choice within the list of all the IPa-programs, either by clicking here [login], or by using the login button at the top-right hand corner of this page.

Since 2014, the field of AI and HEP has grown exponentially. Physicists have realised quickly the potential of AI to deal with the large amount of complex data they are collecting and analysing. Many AI techniques have been put forward, with scientific collaboration based on open data sets, challenges, workshops and papers.


Learning To Discover is a first-of-its-kind workshop where participants will have access to deep technical insights in advanced machine learning techniques, and their application to particle physics. During the event, blending the concept of hackathon, hands-on and tutorials, physicists who have attempted to apply machine learning to specific challenges in HEP will expose their problem case and the solutions they have arrived at so far. ML experts will expose the latest advances in relevant techniques. Machine learning experts and ml-aware physicists will work hand in hand on existing datasets, building upon and improving existing solutions. Under this particular environment, the participants will be able to discuss and understand shortcomings of existing solutions and develop novel architectures and methods to outperform on the specific problem. For example, despite promising applications of Graph Neural Network (GNN) to several HEP problems (tracking, identification, calorimeter reconstruction, particle flow, …) there remains several bottlenecks in the models being used. In particular the following topics are of interest: ways to extract graph-level information, mechanisms for information propagation over the graph, accurate graph generation,  evolution of graph topology within the model, ...

Physicists with concrete experience with Machine Learning (postdocs, advanced PhD students, ...) and Computer Scientists seasoned with the development of cutting edge ML models (expected participation from Google Deepmind, NVidia, IDIAP, ...) are invited to apply for attendance. Application is however opened to everyone with experience and interest. The selection process will be as inclusive as possible, and only done so as to make the workshop fruitful for everyone and given the constraints of the premises.  The final conference on AI and HEP is open to physicists and Computer Scientists until the maximum attendance is reached.

Three themes have tentatively been selected based on the one hand, their interest for HEP, and the fact there is already a number of HEP teams working on it, on the other hand, their importance in the Machine Learning field. The candidate themes to be tackled sequentially are : Graph Neural Networks, Generative Models, and multi dimensional Probability Density Estimation.