Prerequisites: Modules in Mathematics I to IV, basic knowledge in Python.
Exam: oral examination (20 minutes), 5 credit points.
Contact: julien dot vitay at informatik dot tu-chemnitz dot de.
Language: English. The oral exam can of course be done in German.
Note: there is no exam in the summer semester.
The course will present an introduction to the research field of Machine Learning, including Supervised Learning, Deep Learning and Reinforcement Learning. The different algorithms presented during the lectures will be studied in more details during the exercises, through implementations in Python.
The plan of the course is:
- Supervised learning
- Linear machines
- Learning Theory
- Neural Networks
- Support vector machines
- Deep Learning
- Recurrent neural networks
- Reinforcement Learning
- Formal definition of the RL-Problem
- Dynamic Programming
- Monte Carlo Methods
- Temporal Difference Learning (TD)
- Eligibility traces
- Function approximation
- Deep Reinforcement learning
- "Deep Learning", Goodfellow, Bengio & Courville
- "Support Vector Machines and other kernel-based learning methods", Cristianini & Shawe-Taylor
- "Reinforcement Learning", Sutton & Barto (1st and 2nd editions)
- How do I register for the course?
You don't. There is no OPAL or anything, just show up to the lectures/exercises and register for the exam in December.
- How do I register for the exam?
Registration on SBService happens in December. Only registered students can participate to the exam.
- When will the exams be?
The given appointments are at the end of this page. Do not write emails to get another one.
- The appointment I was given does not suit me / I am not ready. Can I have another one?
- How do I unregister from the exam?
Up to one week before your appointment, you simply cancel your registration on SBService AND write me an email so that I do not wait for you in vain. Less that one week before the appointment, it is not possible anymore (like for a written exam), unless you bring a medical certificate to the examination office.
- I cannot assist to the exercises. Can I take the exam anyway?
Yes. The exercises are there to help you understand the concepts seen in the lectures and get practical experience with machine learning tools. But they are not important for the exam.
- Do I have to memorize all these equations?
No, but to understand them, which is basically the same.
- I failed the exam in the winter semester. Can I retake it in the summer semester, otherwise I will not be able to finish my Master thesis?
No, exams are in the winter semester only, no exception. Come prepared to the exam if you are under time pressure.
Slides for the lecturesChapter 01 - Introduction (pdf)
Chapter 02 - Linear learning machines (pdf)
Chapter 03 - Learning theory (pdf)
Chapter 04 - Neural networks (pdf)
Chapter 05 - Support-vector machines (pdf)
Chapter 06 - Deep learning (pdf)
Chapter 07 - Recurrent neural networks (pdf)
Chapter 08 - Reinforcement Learning (pdf)
Chapter 09 - Deep Reinforcement Learning (pdf)
ExercisesExercise 01 - Introduction to Python and NumPy. (pdf , data , solution )
Exercise 02 - Linear classification. (pdf , data , solution )
Exercise 03 - Cross-validation. (pdf , data , solution )
Exercise 04 - Multi-layer perceptron. (pdf , data , solution )
Exercise 05 - Multi-layer perceptron on the MNIST dataset. (pdf , data , solution )
Exercise 06 - Support-vector machines. (pdf , data )
Exercise 07 - Convolutional neural networks. (pdf , data , solution )
Exercise 08 - Transfer learning. (pdf , data , solution )
Exercise 09 - Reinforcement learning. (pdf , solution )
Exercise 10 - Q-learning and Gridworld. (pdf , data , solution )