Formal methods and machine learning both have a long history and a vibrant research community, but they have rarely intersected. Only recently has there been increased interest in developing formal methods for data-driven tasks and applying them to machine-learning-based systems.
The seminar will provide brief introductions into the foundations of both areas, followed by in-depth lectures of state-of-the-art approaches from the literature. The presentations will be given by the participating students. The talk topics will be given out in a pre-meeting in December, with preference given to students who take the course for credit.

Target group: 1) Students and postdocs in computer science who plan to specialize on the intersection of formal methods and machine learning.
2) Students and postdocs in computer science and related areas who are interested in the state of the art in that field

Prerequisites: Required for all participants: good knowledge of computer science.
Required for participants who plan to give a talk (and good to have for all others): good prior knowledge of at least either formal methods or machine learning.

Evaluation: For ECTS credit, participants need to give a talk and participate in the Q&A/discussion.

Teaching format: Seminar format, student present recent research papers in in-depth (60min) presentations, followed by a question/answers/discussion session.

ECTS: 3 Year: 2019

Track segment(s):
CS-AI Computer Science - Artificial Intelligence
CS-SYS Computer Science - Software Systems
DSSC-PROB Data Science and Scientific Computing - Probabilistic Models

Teacher(s):
Krishnendu Chatterjee Christoph Lampert Anna Lukina

Teaching assistant(s):

If you want to enroll to this course, please click: REGISTER