Universität Wien

052323 VU Probabilistic Artificial Intelligence (2026S)

3.00 ECTS (2.00 SWS), SPL 5 - Informatik und Wirtschaftsinformatik
Prüfungsimmanente Lehrveranstaltung
Moodle
Mo 01.06. 13:15-14:45 Seminarraum 7, Währinger Straße 29 1.OG

An/Abmeldung

Hinweis: Ihr Anmeldezeitpunkt innerhalb der Frist hat keine Auswirkungen auf die Platzvergabe (kein "first come, first served").

Details

max. 25 Teilnehmer*innen
Sprache: Englisch

Lehrende

Termine (iCal) - nächster Termin ist mit N markiert

  • Montag 02.03. 13:15 - 14:45 Digital
  • Montag 09.03. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 16.03. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 23.03. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 13.04. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 20.04. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 27.04. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 04.05. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 11.05. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 18.05. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 08.06. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 15.06. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 22.06. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG
  • Montag 29.06. 13:15 - 14:45 Seminarraum 7, Währinger Straße 29 1.OG

Information

Ziele, Inhalte und Methode der Lehrveranstaltung

This course is about building systems that can perform well in uncertain situations without prescribing the optimal behavior. To this end the course covers the fundamentals as well as state-of-the-art methods for building probabilistic artificial intelligence systems like Bayesian modeling, variational inference, generative models (variational auto encoders, diffusion models), and reinforcement learning.
Programming assignments complement the lectures. Upon successful participation in the course, students will understand the fundamentals of probabilistic artificial intelligence, know how to apply basic methods in theory and practice, and will have the ability to connect to current research.

Art der Leistungskontrolle und erlaubte Hilfsmittel

* Written exam (individual work): at the end of the semester; you will be allowed to bring a handwritten A4 sheet (2 pages) of notes.
* Programming assignments (individual or group work): you will solve programming assignments at home; you will have to submit your executable source code and a written report describing the results obtained with your implementation.

Mindestanforderungen und Beurteilungsmaßstab

Requirements.
Solid knowledge in mathematics, probability theory, and algorithms is required to take the course. Solid coding skills are expected. Experience with deep learning frameworks like PyTorch or TensorFlow is a strong plus. You should also have taken an introductory course to machine learning or artificial intelligence so that you know for instance the basics of supervised learning and probabilistic modeling.

Assessment.
Your grade will depend on your performance on the written exam and the programming exercises according to the following weighting:
60% Written exam
40% Programming exercises

In particular, let P = Average weighted percentage on the written exams and the programming exercises. Then your grade (if you fulfill the passing criteria) is given as:
90% <= P <= 100% Sehr Gut (1)
77% <= P < 90% Gut (2)
65% <= P < 77% Befriedigend (3)
50% <= P < 65% Genügend (4)
0% <= P < 50% Nicht Genügend (5)

To successfully complete the course, you need to achieve
* at least 50% of the points on the written exam, AND
* at least 50% of the points on the programming exercises.

If you do not achieve at least 50% of the points on the written exam at the end of the semester, you can take a compensation exam on July 21, 2026 (this might be a written or oral exam, depending on the demand and decided by the lecturer). The compensation exam can only be taken if the 50% requirement is not met. On the compensation exam, you again need to achieve at least 50% of the possible points to satisfy the passing requirements for the course (in addition to the requirements regarding the programming exercises).

Attendance of the lecture parts of the course is voluntary (except for the first lecture for which attendance is mandatory) but highly recommended. Attendance of the written exam is compulsory to pass the course.

Prüfungsstoff

All topics covered in class, the reading material, and the exercises. Referenced literature (as indicated in detail on the lecture slides).

Literatur

Andreas Krause, Jonas Hübotter, "Probabilistic Artificial Intelligence", https://arxiv.org/abs/2502.05244

Zuordnung im Vorlesungsverzeichnis

Letzte Änderung: Fr 20.03.2026 12:26