Topic outline

  • General

    Gaussian processes (GPs) are a powerful machine learning paradigm for Bayesian nonparametric modelling. This course will give an overview of Gaussian processes in machine learning, and it provides both a theoretical and practical background for leveraging them. The course covers Gaussian process regression, classification, and unsupervised modelling, as well as a selection of more recent specialised topics.

    Lecturers: Prof Arno Solin, Dr Markus Heinonen, Prof Aki Vehtari, Dr Martin Trapp, Dr Ti John, and Dr Aidan Scannell

    Teaching assistants: Severi Rissanen and Prakhar Verma

    Prerequisites: Basics of machine learning and statistics, e.g. "Machine Learning: Supervised Methods (CS-E4710)"

    Target audience: The course is targeted towards M.Sc. students interested in deepening their machine learning knowledge:
    • GPs are a probabilistic counterpart of Kernel Methods (CS-E4830)
    • GPs offer a probabilistic way to do Deep Learning (CS-E4890)
    • GPs fall under the umbrella of Bayesian Data Analysis (CS-E5710)
    • GPs utilize Advanced Probabilistic Methods (CS-E4820)

    Format: This is 5-credit course with 12 lectures and 6 home assignments. Lectures:
    • Mondays 10:15–12:00 in Hall U4 (U142)
    • Tuesdays 10:15–12:00 in Hall D (Y122)


    First lecture:     Monday Feb 27th, 10.15, Hall U4 (U142)


    Exercise sessions: The practical assignments will be Python notebooks that are completed at home, returned weekly and graded by course assistants. Other languages (such as Matlab or R) can be used, but it will require more work from the participants. Note that participation awards points that contribute towards the final grade:

    • H1: Thursdays 10:15–12:00 in U142

    We have a Zulip chat which primarily acts as a platform for students to discuss and share interesting articles, resources, etc. Help from teaching assistants is available mainly during the exercise sessions. Link: https://gp-2023.zulip.aalto.fi/join/abccubfoc5z5b7erbyyz7doc/

    Exam: No exam

    Grading: At the end of the course students receive a grade (1 to 5 based on points (max points 48 points, 24 point minimum to pass):
    • Six assignment rounds, each worth 6 points (max 36 points)
    • Two points for participation per weekly exercise session (max 12 points)
    • Two bonus points for returning the course feedback

    Grading table:
    • 1/5, 24– points
    • 2/5, 28– points
    • 3/5, 32– points
    • 4/5, 36– points
    • 5/5, 40– points

    Note: The maximum grade of 5/5 is only possible by attending exercise sessions.

    Book: Rasmussen & Williams (2006). Gaussian Processes for Machine Learning, MIT Press (publicly available).