Uni-Logo

Lectures

Python Workshop from 14. - 16. October 2019

Lecturers:

Prof. Dr. Thiemann (Uni Freiburg)

  • Testing
  • Continuous Integration
  • Git (Hub)


Maximilian Nöthe (TU Dortmund)

  • Scientific Programming
  • Numpy
  • Matplotlib
  • Scipy


Lukas Heinrich (CERN)

  • Symbolic Programming
  • Autodifferentiation
  • Cython/Numba
  • MPI

For further information, follow this link:

https://indico.cern.ch/event/846501/

 

Special lecture by Prof. Lance Dixon from 3. - 4. April 2019

 

"Jets and Energy Correlations in QCD"

Lecture I,  Wednesday, April 3rd, HS I from 14:00 - 15:00h

Lecture II,  Thursday, April 4th, HS I  from 14:00 - 15:00h

Abstract:

I give a pedagogical introduction to the computation of infrared-safe observables in QCD, including jets at hadron colliders, and the energy-energy correlation (EEC) in electron-positron annihilation. Jet observables are ubiquitous at the LHC and can be computed at NLO for a large number of final state jets, but at NNLO for only two jets. On the other hand, the computations are quite numerically intensive. In contrast, the EEC is simple enough that it can be computed fully analytically at NLO, and certain resummations of it can also be performed rather accurately. Some of the tools developed for the EEC might find their way back to the LHC in the guise of a more accurate theoretical understanding of jet substructure.

Slides Part I

Slides Part II

 

Special lecture by Prof. Dr. Gregor Kasciezka
from 9. - 10. April 2018

 

"Basics of Deep Learning & Applications in Particle Physics"

Lecture I, Monday, April 9th, HS I from 09:30 - 12:00h  

Lecture II, Monday, April 9th, HS I from 15:00 - 17:30h  

Lecture III, Monday, April 9th, HS I from 09:30 - 12:00h   

External link to slides

 

Abstract:

Deep learning algorithms nowadays outperform humans in image classification tasks, beat them at challenging games such as Go or Poker, respond to spoken questions, and might soon drive autonomous cars. These achievements were made possible by faster hardware and by using the parallel processing capabilities of modern graphical processing units (GPUs), but at least equally important were developments in software.

Deep learning is the application of many-layered (deep) artificial neural networks. We will start by discussing the essential building blocks, such as activation functions, loss functions, back propagation and gradient descent using different optimiser algorithms. A tour of different deep learning approaches used for classification tasks in particle physics provides an overview of common network architectures. We address strategies how networks can be trained directly on data to rely less on computer simulations and how one can achieve a deeper understanding of the decision process inside the network. Useful concepts such as generative adversarial networks and auto encoders are discussed. Finally, we look at software libraries for implementing deep neural networks.

 deepcms2.jpg

 

 

 

 

 

Special lecture by Prof. Dr. Markus Schumacher
from 26.-28. September 2017

 

"Confidence Intervals and Limits for Pedestrians"

Lecture I, 26. September 2017, HS I from 14:00 - 16:00h  (Slides Part I.pdf)

Lecture II, 27. September 2017, HS I from 14:00 - 16:00h  (Slides Part II.pdf)

Lecture III, 28. September 2017, HS I from 14:00 - 16:00h  (Slides Part III and IV.pdf)

Lecture IV, 29. September 2017, HS I from 13:00 - 15:00h

  

Abstract:

Confidence intervals and its one-sided version called limits try to make a probabilistic statement connecting the outcome of an experiment for an unknown parameter and its true value. Frequentist and Bayesian intervals, despite agreeing for many use cases, have a completely different meaning. The concept of constructing the different intervals and their interpretation will be discussed. Differences appear close to physical boundaries e.g. signal yields or mass values close to zero. In this context alternative concepts as the power constrained limits and the CL_s technique will be described and compared to traditional frequentist and Bayesian limits. Finally the flip-flop feature and a possible solution in the unified approach yielding so-called Feldman-Cousins limits will be exemplified. The lecture requires no specific pre-knowledge in statistics and only simple examples based on the Poisson and Gauss probability distributions will be discussed.

Limits Gauss, lecture by M. Schumacher

 

 

 

Special lecture by Felix Kahlhoefer (DESY) and Marc Schumann (Uni Freiburg)


"Dark Matter (Theory and Experiment)" from April 19 - 21, 2017
 

- Slides Felix Kahlhoefer (external link)

 

 

 

Special Lecture by Dr. Markus Elsing from April 12 - 13, 2016 

Markus Elsing
"Tracking at the LHC"

 

Lecture I, 12. April 2016, HS II from 14:00 - 16:00h

Lecture II, 13. April 2016, HS II from 09:30 - 11:30h

 

Abstract:

Event reconstruction for the Large Hadron Collider (LHC) experiments is posing many unprecedented challenges. The LHC is a high luminosity machine operating at centre of mass energies of up to 13 TeV, producing in excess of 30 proton-proton interactions in each bunch crossing at a frequency of 40 MHz. The reconstruction of all trajectories of the charged particles produced in those interactions is not only extremely computationally expensive, but it as well requires very sophisticated techniques to reach the precision and purity require to enable the LHC physics program.

Illustration zur Vorlesung Elsing

In the lecture I will give an introduction to tracking and vertexing at the LHC. I will give an introduction on how the physics processes for particles passing through matter affects the expected tracking performance, followed by a brief overview of the LHC detectors. I will then discuss reconstruction concepts and techniques that are used to master the tracking challenge at the LHC. I will explain how track propagation in a realistic detector works, present different techniques for track fitting and track finding. I will as well cover vertex reconstruction techniques, including their application to b-tagging and pileup mitigation. I will conclude with an outlook on the ATLAS and CMS upgrade plans for Phase-2 that both feature all new tracking detectors designed for the challenges of High Luminosity LHC.

 

Link to the slides of the lecture 

 

 

Special Lecture by Prof. Dr. Norbert Wermes from
April 6 - 8, 2016

     

"Tracking Detectors Lecture"
 

Lecture I,  06. April 2016, Großer Hörsaal Physik from 14:00 - 15:00 h  

Lecture II,  07. April 2016, Großer Hörsaal Physik from 09:00 - 10:00 h  

Lecture III,  07. April 2016, Großer Hörsaal Physik from 14:00 - 15:00 h  

Lecture IV, 08.  April 2016, Großer Hörsaal Physik from 09:00 - 10:00h
 

Abstract:

The three lectures will cover basics and advanced topics on "particle tracking and tracking detectors".
 
The first lecture will give basics on "tracking": how to determine a track or a secondary vertex and which 
are important ingredients for good quality measurements. The Gluckstern formula will be derived.
 
Lecture two deals with the basics of signal development by induced charges in parallel and cylindrical configurations as well as space point reconstruction. What is better, center-of-gravity method or eta-algorithm? This lecture also covers basics of wire chambers as tracking detectors and recent advances.
 
Lecture three covers semiconductor tracking detectors, radiation damage and advanced pixel detectors.

 

Link to the slides of the lecture  

 
 
 

Special Lecture by Prof. Dr. Stefan Dittmaier from
October 13 - 15, 2015

     

"Electroweak Physics at the LHC"
 

Lecture I,  13. October 2015, Physics Highrise, HS II, 10 ct  (slides)
Lecture II, 14. October 2015, Physics Highrise, HS II,  10 ct   (slides)
Room changement:  Lecture III, 15. October 2015, Gustav-Mie-Haus, Seminar room,  10 ct (slides)

 

Benutzerspezifische Werkzeuge