Introduction to Machine LearningThe course will introduce the foundations of learning and making predictions from data. We will study basic concepts such as trading goodness of fit and model complexity. We will discuss important machine learning algorithms used in practice, and provide hands-on experience in a course project. VVZ Information is available here.
- EASTER BREAK: There will be no instructor Q&A on April 14&15 and no tutorials on April 15.
- Starting now, the instructor Q&A is now recorded as well. The files are password-protected, please use the same credentials as you would use for accessing the 2020 slides (see last point of news).
- Due to the complete digitalization of the lecture, we will no longer do the lectures in the usual format. Please watch last year’s videos, which you can access using your normal nethz credentials. During lecture time, we will instead hold zoom meetings, where you can ask questions directly to Andreas Krause. For more details, please refer to your personal email or to piazza.
- Due to Corona, there will be no more physical lectures and office hours. Lecture recordings will still be available on the ETH video portal, for questions, please refer to piazza or the tutorial Q&A. Also, there will be no more public viewing of the tutorials, please use your own zoom client.
- The video recordings are now available at the ETH Videoportal. Password is the same as for the documents on this website. To obtain the password you need to be inside the ETH network and click here. To establish a VPN connection click here.
- The exam will take place on a computer. Please inform us about any special request regarding a disability.
- In the first week’s tutorial sessions on 19. February, we will offer a review session of required background material for the course. This will include a short recap of linear algebra, multivariate analysis and probability theory.
- For programming background, we recommend knowing Python. For those without experience in it, check one of the excellent tutorials Python Tutorial I or Python Tutorial II
- For the mathematical background check the excellent resource: Mathematics for Machine Learning Online version
- The files are password protected. To obtain the password you need to be inside the ETH network and click here. To establish a VPN connection click here.
- 1. If this course is compulsory for your study program (Kernfach), you are able to register irrespective of the waiting list. Please allow some time for the transfer from the waiting list.
- 2. The project is a mandatory part of the examination. Without achieving a passing grade (4), you are not allowed to sit the final examination.
- 3. Distance examination is allowed but you need to file an official request via study administration. We do not handle these requests.
- 4. Attendance at Tutorials and Lectures is not mandatory.
Date Topics Tutorial HW Q&A Tue 18.2. Introduction – – – Wed 19.2. Regression Math Recap HW 1 – Tue 25.2. Generalization and Validation – – – Wed 26.2. Model Selection Regression (black), white, supp, code Sol 1 – Tue 3.3. Classification – – – Wed 4.3. Feature Selection Classification HW 2 – Tue 10.3. Kernels I – – – Wed 11.3. Kernels II Kernels, code Sol 2 – Tue 17.3 Kernels III - - Q&A Wed 18.3 Class Imbalance Kernels 2 HW 3 Q&A Tue 24.3 Neural Networks – – Q&A Wed 25.3 Neural Networks II Imbalance Sol 3 Q&A , video Tue 31.3 Neural Networks III – – – Wed 1.4. Neural Networks IV – – –
Video LecturesThe video recordings of the lectures and tutorials are available at the ETH Videoportal. The files are password protected. To obtain the password you need to be inside the ETH network and click here. To establish a VPN connection click here.
|Instructors||Prof. Andreas Krause|
|Head TA||Philippe Wenk|
|Assistants||Andisheh Amrollahi, Nemanja Bartolovic, Ilija Bogunovic, Zalán Borsos, Charlotte Bunne, Sebastian Curi, Radek Danecek, Gideon Dresdner, Joanna Ficek, Vincent Fortuin, Carl Johann Simon Gabriel, Shubhangi Gosh, Nezihe Merve Gürel, Matthias Hüser, Jakob Jakob, Mikhail Karasikov, Kjong Lehmann, Julian Mäder, Anastasia Makarova, Mojmír Mutný, Gabriela Malenova, Harun Mustafa, Mohammad Reza Karimi, Max Paulus , Laurie Prelot, Jonas Rothfuss, Stefan Stark, Jingwei Tang, Xianyao Zhang,|
|Piazza||If you have any questions, please use the Piazza Course Forum.|
|Mailing List||Please use the Piazza Forum for questions regrading course material, organisation and projects. If this does not work for your request, please use the tutorial webinar to ask questions.|
|There will be no more life lectures. Instead, there will be Q&A sessions with Andreas Krause via zoom to discuss the lecture material.|
|Tutorials will be conducted as a live webinar. For more information, please refer to piazza, your personal email or the lecture slides of the first week. The webinar will take place every Wednesday, starting at 15:15 and lasting for about two hours.|
Problem SetsHomeworks will be distributed electronically and partially graded on the moodle platform . They are intended for you to practice concepts and your performance in the homeworks will in no way affect your final grade. They are published bi-weekly, with solutions following one week after or being directly visible after entering your solutions in moodle.
ProjectPlease find all information about the project here.
DemosThe Demo’s are based on jupyter notebook (with python 3). Please look at this intro for installing and running instructions. We also provide a README File in case you want to install a conda environment where to run the demos. Here are also the requirements.txt for the environment.
Helper files: (Please download them and save inside a ‘utilities’ folder located in the same directory as the demos.). zipped utilities (Updated 18.03.20)
- Linear Regression and Gradient Descent
- Linear Regression on Non-Linear Features
- Gradient Descent on Other Functions
- Model Selection with Sklearn
- Regularization and the Bias Variance Trade-Off
- Perceptron and SVM Demo
- Feature Selection
- Class Imbalance
- Multi-Class Classification
- Multi-Class Classification
- Neural Networks
Performance Assessment70% session examination, 30% project; the final grade will be calculated as weighted average of both these elements. As a compulsory continuous performance assessment task, the project must be passed on its own. The exam might take place at a computer. The practical projects are an integral part (60 hours of work, 2 credits) of the course. Participation is mandatory. Failing the project results in a failing grade for the overall examination of Introduction to Machine Learning (252-0220-00L). Students who do not pass the project are required to de-register from the exam and will otherwise be treated as a no show.
For the final exam, you can bring two A4-pages (i.e. one A4-sheet of paper), either handwritten or 11 point minimum font size. No calculators or other aids are allowed.
- Marc Peter Deisenroth, A Aldo Faisal, and Cheng Soon Ong Mathematics for Machine Learning Online version
- K. Murphy. Machine Learning: a Probabilistic Perspective. MIT Press 2012
- C. Bishop. Pattern Recognition and Machine Learning. Springer, 2007 (optional)
- T. Hastie, R. Tibshirani, and J. Friedman. The Elements of Statistical Learning: Data Mining, Inference and Prediction. Springer, 2001. Available online
- L. Wasserman. All of Statistics: A Concise Course in Statistical Inference. Springer, 2004.