Fundamentals of Statistical Pattern Recognition

Posted on Tue 13 April 2021 in courses

EPFL's logo

This course (EE-612) presents fundamental tools used in Machine Learning and Pattern Recognition ranging from the most basic to more advanced (e.g. Logistic Regression, Principal Component Analysis, Linear Discriminant Analysis, Multi-Layer Perceptrons, Deep Learning, Gaussian Mixture Models, and Super Vector Machines). This course can serve as a pre-requisite for more advanced course on Machine Learning. It is given to post-grad (Ph.D.) students at the École Polytechnique Fédérale de Lausanne, Switzerland.

Outcomes: this course provides in-depth understanding in Machine Learning as well as concrete tools to PhD students for their work. This course could serve as a pre-requisite for more advanced courses such as Machine Learning, Graphical Models, Statistical Sequence Processing and Computational perception using multimodal sensors.

This course has been well received by our students so far:

  • 2012-2013: 7 students, no evaluations
  • 2014-2015: 18 students, 12 evaluations (grade 5.4/6.0)
  • 2016-2017: 24 students, 24 evaluations (grade 5.2/6.0)
  • 2018-2019: 19 students, 19 evaluations (grade 5.5/6.0)
  • 2020-2021: 24 students, ? evaluations (grade ?.?/6.0)

Program:

  • Credits: 4 ECTS (equivalent to 100 to 120 working hours)
  • Grading: Lab assignments (40%) and Final exam (60%)
  • Days: Thursdays, from 10:15 to 14:00 with a break in the middle
  • Required prior knowledge: Linear algebra, Probabilities and Statistics, Python Programming
  • Syllabus:
    • Introduction to Machine Learning
    • Linear Regression
    • Logistic Regression
    • k-Neareast Neighbours
    • Decision Trees
    • Boosting
    • Dimensionality Reduction:
      • Principal Component Analysis (PCA)
      • Linear Discriminant Analysis (LDA)
      • t-distributed Stochastic Neighbour Embedding (t-SNE)
    • Probability distribution modelling:
      • k-Means
      • Gaussian Mixture Models (GMM) and Expectation-Maximimization (EM)
    • Multi-layer Perceptrons and Gradient Descent
    • Deep Learning:
      • Basics
      • Optimisation
      • Architectures
    • Support Vector Machines

Resources:

  • EPFL coursebook: contains the course syllabus and room locations for various days
  • Moodle page: contains slides for all courses and lab assignments (N.B.: enrollement required)
  • JupyterHub: Course assignments are provided as Jupyter Notebooks through JupyterHub (no software installation required on student laptop)