FYI
Begin forwarded message:
From: LOPEZ HONOREZ Laura <Laura.Lopez.Honorez@ulb.be mailto:Laura.Lopez.Honorez@ulb.be> Subject: program of Lectures@PhysTH: Intro ML methods Date: 18 September 2023 at 16:33:03 CEST To: GRP_ULB_PhysTh <GRP_phys_th@ulb.be mailto:GRP_phys_th@ulb.be> Cc: Bryan Zaldívar <bryan.zaldivar@ific.uv.es mailto:bryan.zaldivar@ific.uv.es> Resent-From: <alberto.mariotti@vub.be mailto:alberto.mariotti@vub.be>
Dear all,
here is some more info on the Intro ML methods lectures that will take place in Solvay room at ULB (NO building 5th floor) on 2-5/10, see below and on the indico webpage https://indico.iihe.ac.be/event/1805/ (the Tuesday-Thursday lectures can be found clicking on 2-4 in the menu "From the same series").
Notice that we have added two hours on Monday!
Everyone is welcome and registration is open!
Best,
Laura
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Lectures@PhysTH: Introduction to Machine Learning methods
Lecturer: Bryan Zaldivar
Theoretical Physicist https://inspirehep.net/authors/1239220?ui-citation-summary=true and expert in applying machine learning methods https://aiinitiative.science/people-institutions/ to HEP physics problems and beyond.
Abstract: The main scope of these lectures is to provide the attendee with the basic mathematical/statistical knowledge to understand and use Machine Learning tools in her/his work. It is also conceived to be useful for practical implementations in Python language. The main focus will be on "Supervised Learning" techniques, considering both frequentist and Bayesian approaches. The format is mostly blackboard-based, while sometimes python codes will be shown. A basic background on statistics would be beneficial from the attendees, although the lectures aim at starting at a very elementary level. Scheduled per day (Solvay room) :
Monday 2/10: 10-12h: 1. Overview of ML. 2. summary of statistics
Tuesday 3/10: 10-12h: 3. Regression & overfitting control 4. Bayesian learning
Wednesday 4/10: 15-17h 5. Classification 6. Neural networks
Thursday 5/10: 14-16h 7. To be defined, depending on the interest: Hands-on session with Python on previous methods, or Non-parametric methods, including Gaussian Processes (theory), or Unsupervised Learning methods (theory)
-- Laura Lopez Honorez Service de Physique Théorique ULB, CP 225 Boulevard du Triomphe, 1050 Bruxelles, Belgium
office: 2N7 113 tel: 0032 (0)2 650 55 19