Data Analysis Block Course: Statistics and Machine learning

Name: Data Analysis Block Course: Statistics and Machine learning
Start: 2026-07-20T09:00:00+02:00
End: 2026-08-10T13:00:00+02:00
Location: TUD/RUB/Hybrid

20 Jul 2026, 09:00 → 10 Aug 2026, 13:00 Europe/Zurich

TUD/RUB/Hybrid

TUD: Otto-Hahn-Straße 4a, 44227 Dortmund, CP-03-123 RUB: Universitätstr. 150, 44780 Bochum, NB2/158

Description

(curerntly, the content of the course is clonned from our 2025 ML course, and subject for adjustment)

Literature:

G. Cowan , Statistical Data Analysis
O. Behnke et al., Data Analysis in High Energy Physics

Previous courses:

RUB course, (repository)
TUD 2024

Monday 20 July
- Mon 20 Jul
- Tue 21 Jul
- Wed 22 Jul
- Thu 23 Jul
- Mon 27 Jul
- Tue 28 Jul
- Wed 29 Jul
- Thu 30 Jul
- Mon 10 Aug
- 1
  
  Lecture - Introduction
  
  Speaker: Michael Schmelling (Max Planck Society (DE))
- 2
  
  Exercises
  
  Speaker: Jan Peter Wagner (Technische Universitaet Dortmund (DE))
- 3
  
  Lecture - Error propagation
  
  Speaker: Michael Schmelling (Max Planck Society (DE))
- 4
  
  Exercises
  
  Speaker: Jan Peter Wagner (Technische Universitaet Dortmund (DE))
Tuesday 21 July
- Mon 20 Jul
- Tue 21 Jul
- Wed 22 Jul
- Thu 23 Jul
- Mon 27 Jul
- Tue 28 Jul
- Wed 29 Jul
- Thu 30 Jul
- Mon 10 Aug
- 5
  
  Lecture - Information, inference and confidence level intervals
  
  Speaker: Michael Schmelling (Max Planck Society (DE))
- 6
  
  Exercises
  
  Speaker: Jan Peter Wagner (Technische Universitaet Dortmund (DE))
- 7
  
  Lecture - Least Squares Fitting
  
  Speaker: Michael Schmelling (Max Planck Society (DE))
- 8
  
  Exercises
  
  Speaker: Jan Peter Wagner (Technische Universitaet Dortmund (DE))
Wednesday 22 July
- Mon 20 Jul
- Tue 21 Jul
- Wed 22 Jul
- Thu 23 Jul
- Mon 27 Jul
- Tue 28 Jul
- Wed 29 Jul
- Thu 30 Jul
- Mon 10 Aug
- 9
  
  Lecture - Maximum Likelihood, Sandwich estimator, Bootstrap
  
  Speaker: Michael Schmelling (Max Planck Society (DE))
- 10
  
  Exercises
  
  Speaker: Jan Peter Wagner (Technische Universitaet Dortmund (DE))
- 11
  
  Lecture - Markov Chain Monte Carlo, COWs
  
  Speaker: Michael Schmelling (Max Planck Society (DE))
- 12
  
  Exercises
  
  Speaker: Jan Peter Wagner (Technische Universitaet Dortmund (DE))
Thursday 23 July
- Mon 20 Jul
- Tue 21 Jul
- Wed 22 Jul
- Thu 23 Jul
- Mon 27 Jul
- Tue 28 Jul
- Wed 29 Jul
- Thu 30 Jul
- Mon 10 Aug
- 13
  
  Lecture - sPlot, sFit, Unfolding
  
  Speaker: Michael Schmelling (Max Planck Society (DE))
- 14
  
  Exercises
  
  Speaker: Jan Peter Wagner (Technische Universitaet Dortmund (DE))
- 15
  
  Lecture - Unfolding, Harmonic analysis
  
  Speaker: Michael Schmelling (Max Planck Society (DE))
- 16
  
  Exercises
  
  Speaker: Jan Peter Wagner (Technische Universitaet Dortmund (DE))
Monday 27 July
- Mon 20 Jul
- Tue 21 Jul
- Wed 22 Jul
- Thu 23 Jul
- Mon 27 Jul
- Tue 28 Jul
- Wed 29 Jul
- Thu 30 Jul
- Mon 10 Aug
- 17
  
  Automatic Differentiation
  
  Speaker: Prof. Mikhail Mikhasenko (Ruhr Univeristy Bochum)
- 18
  
  Exercises - Automatic Differentiation
  
  Exercise notebook
  
  Solution notebook
- 19
  Machine Learning (loss, ROC)
  - Slides
  - lecture-1B-entropy.jl: introduction to entropy
  - lecture-1B-mse.jl: MSE as the simplest loss function
  Speaker: Prof. Mikhail Mikhasenko (Ruhr Univeristy Bochum)
- 20
  
  Exercises - Samples, FoM, ROC
  
  Exercise notebook
  
  Solution notebook
Tuesday 28 July
- Mon 20 Jul
- Tue 21 Jul
- Wed 22 Jul
- Thu 23 Jul
- Mon 27 Jul
- Tue 28 Jul
- Wed 29 Jul
- Thu 30 Jul
- Mon 10 Aug
- 21
  Decision Trees, Adaboost
  - Slides
  - lecture-2A-split.jl: splitting index demonstration
  - lecture-2A-trees.jl: decision tree demonstration
  Speaker: Prof. Mikhail Mikhasenko (Ruhr Univeristy Bochum)
- 22
  
  Exercises - Decision Trees
  
  Exercise notebook
  
  Solution notebook
- 23
  Decision Trees, GradientBoost
  - Slides
  - GBDT playground
  Speaker: Prof. Mikhail Mikhasenko (Ruhr Univeristy Bochum)
- 24
  
  Exercises - AdaBoost, GradientBoost
  
  Exercise notebook
  
  Solution notebook
Wednesday 29 July
- Mon 20 Jul
- Tue 21 Jul
- Wed 22 Jul
- Thu 23 Jul
- Mon 27 Jul
- Tue 28 Jul
- Wed 29 Jul
- Thu 30 Jul
- Mon 10 Aug
- 25
  Neural Networks : General introduction
  - Slides
  - lecture-3A-xor.jl: learning xor demonstrator
  - gradient descent app: github.com/lilipads/gradient_descent_viz
  Speaker: Prof. Mikhail Mikhasenko (Ruhr Univeristy Bochum)
- 26
  
  Exercises - Neural Networks
  
  Exercise notebook
  
  Solution notebook
- 27
  Neural networks : Bias-Variance TradeOff and regularization techniques
  - Slides: Lecture 3 pdf contains but A,B parts
  - lecture-3B-UAT.jl: Universal approximation theory demonstration
  - lecture-3B-NMIST.jl: training of MNIST classification
  - playground.tensorflow.org
  Speaker: Prof. Mikhail Mikhasenko (Ruhr Univeristy Bochum)
- 28
  
  Exercises - NNs on example data
  
  Exercise notebook
  
  Solution notebook
Thursday 30 July
- Mon 20 Jul
- Tue 21 Jul
- Wed 22 Jul
- Thu 23 Jul
- Mon 27 Jul
- Tue 28 Jul
- Wed 29 Jul
- Thu 30 Jul
- Mon 10 Aug
- 29
  Generative networks, Normalizing flow
  - Slides
  - lecture-4A-transformations.jl: demonstration of bijectors
  - lecture-4A-RQS.jl: tuning RQS
  Speaker: Prof. Mikhail Mikhasenko (Ruhr Univeristy Bochum)
- 30
  
  Exercises - Normalizing flow
  
  Notebook NF basics
  
  Notebook NF training
- 31
  
  BBQ
  
  Sign up here: https://hackmd.e5.physik.tu-dortmund.de/3XzYG8C4ScuKaNh43QB3GQ?both
  
  Location: https://maps.app.goo.gl/qwZvu8eUB14JzKTw5
Monday 10 August
- Mon 20 Jul
- Tue 21 Jul
- Wed 22 Jul
- Thu 23 Jul
- Mon 27 Jul
- Tue 28 Jul
- Wed 29 Jul
- Thu 30 Jul
- Mon 10 Aug
- 32
  Written Exam (in person, in Dortmund and Bochum)
  - Pen and paper, calculator (non-programmable)
  - Time: 1:30h
  - 5-6 Problems
  - ChatGPT is not allowed