To achieve state-of-the-art performance with machine learning (ML), users, developers and researchers have to make many design decisions, incl. ML algorithm, pre-processing, post-processing and their hyperparameters. The course on automated machine learning (AutoML) at the ai-campus.org conveys the basics on how to make these design decisions automatically, s.t. ML practitioners can efficiently apply ML to new applications. The course covers e.g. hyperparameter optimization, Bayesian optimization, neural architecture search and meta-learning.
To achieve state-of-the-art performance with machine learning (ML), users, developers and researchers have to make many design decisions, incl. ML algorithm, pre-processing, post-processing and their hyperparameters. The course on automated machine learning (AutoML) at the ai-campus.org conveys the basics on how to make these design decisions automatically, s.t. ML practitioners can efficiently apply ML to new applications. The course covers e.g. hyperparameter optimization, Bayesian optimization, neural architecture search and meta-learning. To this end, the course offers video lectures, quizzes and coding exercises. Furthermore, the course is a joint project by Bernd Bischl (LMU München), Frank Hutter (University Freiburg) and Marius Lindauer (LUH), and supported by Lars Kotthoff (University of Wyoming), Joaquin Vanschoren (Eindhoven University of Technologoy) and Janek Thomas (Fraunhofer SCS).
This allows us to combine the expertise of world-leading researcher in the field of AutoML and to design a unique course that covers both the basics and recent advances. Students from the LMU München, the University of Freiburg and the Leibniz University Hannover (LUH) can attend this course as part of their studies (e.g., CS) and everyone else (e.g., practitioners from industry) can enroll via the ai-campus.org.
This project is funded by the BMBF.
