Because machine Learning can only be understood through practice, by using the algorithms, the course is accompanied with practicals during which students test a variety of machine learning algorithm with real world data . The courses uses matlab libraries for machine learning, as well as the MLDEMOS TOOLBOX that entails a large variety of Machine Learning algorithms.

• Pattern recognition and clustering: PCA, kNN, K-means, DBSCAN.
• Binary and multi-class classifiers: LDA, GMM with Bayes, SVM, Boosting, Neural Networks.
• Non-linear Regression: SVR, GMR, Neural Networks

Machine Learning, Statistics

Linear Algebra, Probability & Statistics

Linear Algebra: Eigenvalue and singular value decomposition

Statistics: Definitions of probability density function, marginal, likelihood, covariance, correlation

Optimization: Lagrange multipliers, gradient descent, local and global optima

• Use a work methodology appropriate to the task.
• Continue to work through difficulties or initial failure to find optimal solutions.
• Write a scientific or technical report.
• Make an oral presentation.

The format of the course is that of a Split-Class: The theory of the course is presented through a video which students must watch prior to class. One hour of the course is allocated for this. This is following by an ex-cathedra lecture that consists of a one-hour interactive lecture session. The interactive lecture takes place on campus, but students who need it can also attend through zoom. An electronic polling system is used to engage students during the lecture.

A two-hour exercise sessions is given each week after the lecture. 4 of the weeks of the course are dedicated to computer-based practical sessions, during which students learn to use the algorithms seen in class for processing real data.

Exercise sessions and computer-based practice session are done on site only. No remote connection possible.

Students are expected to watch the videos prior to the interactive lecture, as the interactive lecture will not repeat the video but go in more depth in the concepts presented in the videos.

Students are expected to attend the exercise sessions and the computer-based practice sessions. They should revise the class notes prior to going to practical session to be on top of the the theoretical material prior to applying it.

Students who are no longer up to date with the pre-requisites should work on these in parralel to taking the class.

Final written exam (100% grade), in-class assessment through a quiz (0% grade).

The course is accompanied with lecture notes entitled: Machine Learning Techniques, available as a pdf posted on the class's website.

• https://go.epfl.ch/MICRO-455https://go.epfl.ch/MICRO-455

• https://tube.switch.ch/channels/7e86d16dhttps://tube.switch.ch/channels/7e86d16d

MICRO-570 Advanced Machine Learning, spring semester

MICRO-462 Learning and adaptative control for robots