Talks/Teaching

 

Conferences

Excluding national events, internal ATLAS collaboration events,  and presentations from co-authors. Most important ones are highlighted.

  • “Physics Based Machine Learning”, David Rousseau, Invited Talk, virtual, Machine Learning for Scientific Imaging at Electronic Imaging 2022, San Francisco, Jan 2022
  • “End to end learning with an Optical Processing Unit“, David Rousseau, Advanced Computing and Analysis Techniques in Physics Research 2021, virtual, Daejeon, South Korea, Nov 2021

  •  “Recent progresses in using Artificial Intelligence for Particle Physics“, David Rousseau, Invited talk, American Physical Society April Meeting, Data Science and Machine Learning in Particle and Astrophysics session, virtual, April 2021

  •  “Using an Optical Processing Unit for tracking and calorimetry at the Large Hadron Collider” David Rousseau,  Fast Machine Learning for Science Workshop, SMU, Dallas (virtual), Dec 2020  

  • “Using an Optical Processing Unit for tracking and calorimetry at the Large Hadron Collider“, David Rousseau,  CERN Interexperiment Machine Learning Workshop, Oct 2020

  • “Novel data analysis techniques”, David Rousseau, plenary talk, European Physical Society Conference on High Energy Physics 2019, Ghent, July 2019

  • “Advances in Machine Learning for HEP”, David Rousseau, plenary talk, Conference on High Energy Physics, Sofia, July 2018 
  • “Organising scientific data challenges”, David Rousseau, Institut Pascal Workshop on Advanced Pattern Recognition, 14-25 Oct 2019 
  •  “Machine Learning in HEP : GAN, TrackML and more ”, David Rousseau, CERN/SKA workshop, Alan Turing Institute, London 17th Sep 2018 
  • ”Advances in Machine Learning in High Energy Physics : Deep Learning, GAN and more” David Rousseau, AGATA/GRETINA tracking arrays meeting, 4-6 April 2018, Orsay 
  •  “LHC experiments : Petabytes to papers”, David Rousseau, LSST school and workshop : getting ready to do science with LSST data, Lyon, 15 juin 2017 
  • Talks about the TrackML challenge:
    • “TrackML : the roller coaster of organizing a HEP challenge on Kaggle and Codalab”, David Rousseau, European Physical Society Conference on High Energy Physics 2019, Ghent, July 2019 
    • “TrackML Conclusion and Outlook on the on-going Throughput phase”, David Rousseau, NeurIPS 2018 Competition Workshop, Montréal,4-8 Dec 2018
    • “TrackML, the Tracking Machine Learning challenge”, David Rousseau, RAPID workshop, Dortmund, 19-21 Nov 2018 
    • “TrackML, the Tracking Machine Learning challenge, and a bit more”, David Rousseau, Advanced Statistics for Physics Discovery, Padova, 25-26th Sep 2018 
    • “Tracking machine learning challenge hackathon”, David Rousseau, Connecting The Dots 2018, Seattle, 20-22 mars 2018 
    • “Tracking machine learning challenge hackathon”, David Rousseau, Connecting The Dots / Intelligent Tracker 2017, Orsay, 6-9 mars 2017
    • “Status of the tracking machine learning challenge”, David Rousseau, Connecting The Dots 2016, Vienna, 22-24 février 2016 
    • “A HEP pattern recognition challenge on Kaggle ? Experience from the ATLAS Higgs Machine Learning challenge”, David Rousseau, Connecting The Dots 2015, Berkeley, 9-11 février 2015 
  •  “ATLAS Machine Learning and Data Analytics challenges”, CERN OpenLab Machine Learning and Data Analytics workshop, 29 April 2016, CERN 
  • “Data Science at LHC Workshop Highlights”, Dirk Duellman and David Rousseau, ”Machine Learning for the LHC Distributed Data Placement and Track Finding”, Moscou, 7-9 Décembre 2015 
  • The Higgs Machine Learning Challenge”, D. Rousseau, C. Adam- Bourdarios, G. Cowan, C. Germain, I. Guyon, et al., Higgs Machine Learning Challenge visits CERN, May 2015, Geneve, Suisse.
  • “Higgs boson physics : experimental results”, D. Rousseau, PANIC, Hambourg, Allemagne, 25-29 août 2014
  • “Software” in session Trigger/DAQ/Offline/Computing” of the ”ECFA High Luminosity LHC Experiments Workshop” Aix-les-bains, 1-3 octobre 2013
  • “Higgs Searches in ATLAS ”, D. Rousseau, High Energy Physics in the LHC era, Valparaiso, Chili, 4-10 janvier 2012. ATL-PHYS-SLIDE-2012-010.
  • “ATLAS software challenges”, D. Rousseau, Future computing in particle physics, e-Science Institute, Edinburgh, 15-17 juin 2011.
  • “The ATLAS Liquid Argon Calorimeter : test beam, installation and commissioning”, D. Rousseau, IEEE 2007 Nuclear Science Symposium and Medical Imaging Conference, Hawaii, 27 oct-3 nov 2007.
  • “LHC detector studies and early physics”, D. Rousseau, 2007 Aspen Winter Conference ”New physics at the electroweak scale and new signals at hadron colliders”, Aspen, 8-13 janvier 2007. ATL-SLIDE-2007-005
  • “ATLAS reconstruction software”, D. Rousseau, EPS-ICHEP 2003, Aachen, Germany 17-23 juillet 2003, Eur. Phys. J. C 33 (2004) S1038. ATL-COM-SOFT-2003-015.
  • “ATLAS B physics overview”, D. Rousseau, Beauty 2000, Kibbutz Ma’agan, Israel, 13-18 septembre 2000, Nucl. Instrum. Meth. A 462, 189 (2001).
  • “Heavy flavours decay at LEP”, D. Rousseau, Moriond QCD and High Energy Interactions, Les Arcs, France, mars 1999, Editions Frontière
  • “CP violation with the ATLAS detector”, D. Rousseau, International Conference on Hyperons, Charm and Beauty Hadrons, Genoa, Italie, Mai 1998, Nucl. Phys. Proc. Suppl. 75B, 351 (1999). ATL-CONF-99-001.
  • “Measurement of Vcb and form factors and branching fractions in the decays B0 → D∗+l−νl and B0 → D+l−νl”, D. Rousseau, Frontiers in Contemporary Physics, Nashville, USA, 1997.
  • “Measurement of Vcb ”, D. Rousseau, Lake Louise Winter Institute on Quarks and Colliders, Lake Louise, Canada, F ́evrier 1995, actes: Edited by A. Astbury, B.A. Campbell, F.C. Khanna, J.L. Pinfold. Singapore, World Scientific, 1996, 542p.

 

Seminars

  • D. Rousseau “AI4HEP” journées DEDIP, CEA-Saclay, April 2022 ; Laboratoire de Physique du Solide, Orsay, Dec 2022
  • D. Rousseau ”Machine Learning in HEP : GAN, TrackML and more”, Oxford Physics Department 24th Oct 2019
  • D. Rousseau ”The TrackML HEP Tracking Machine Learning challenge”, LRI-Orsay Feb 2018; CERN Mar 2018;  Heidelberg May 2018
  • D. Rousseau, ”Progress in Machine Learning Tools for High Energy Physics”, LAL-Orsay May 2016; CEA/IRFU/SPP-Saclay May 2016;  LPNHE-Paris Jun 2016; LPSC-Grenoble Oct 2016; LPC-Clermont Oct 2016; LAPP-Annecy Feb 2017; IHP-Strasbourg Oct 2017;  Uppsala (Sweden) Oct 2017; LLR-Palaiseau Nov 2017; JINR day Centre de Russie pour la Science et la Culture Paris Feb 2018; CEA/DRF May 2018; LPNHE Paris, Nov 2020

Summer schools

I’ve given many courses on the theme “Machine Learning and Particle Physics” in various “summer” / undergrad / grad schools, adapted to the audience, settings (on-site/video/hybrid) and duration (1 to 6 hours):

 

If possible given the format of the school,  I also run a tutorial on the basics of training and evaluating a Boosted Decision Tree or Neural Network classifier on a typical HEP dataset (4-vectors from Higgs to WW events from ATLAS Open Data). The tutorial is on github (developed with Aishik Ghosh and Jérémy Couthures) and can readily be run on Google Colab, and also with minor mods run on one’s laptop with anaconda installation; I however do not recommend its use standalone as it requires background information (on ML, HEP and Higgs physics).

A bit off-topic : at the beginning of ATLAS, many students and experienced physicists as well need to be trained to use the ATLAS framework Athena, Event Data Model and algorithms. I’ve given many tutorials to that respect.

I also gave a course on “Physics at the LHC” at the VIII Mexican School for Particle Physics and Fields, Oaxaca, Mexico, Nov 1998 (AIP Conference Proceedings 490)

Teaching

With Jean-Christophe Hamilton (CNRS APC) and  Guillaume Mention (CEA) I give a course since 2019 on “Statistics and Machine Learning for Cosmology and Particle Physics” to first year (pre-master) CentraleSupelec (Université Paris-Saclay) students. About 120 students who have chosen the Physics and NanoTechnology (PNT) theme. I teach the Machine Learning part (16 hours). In addition to courses and tutorials, we have  one week full-time “Enseignement d’Intégration” where students in “Collaboration” of 25 (5 teams of 5) have to build a complete ML analysis pipeline of an LHC dataset.

In addition

  • “stages d’observation de 3ème” at IJCLab, one week
  • projet recherche  double licence Math Physique Orsay (1/2 day per week for 2 months) in 2020, 2021, 2023