Mathematical Structures of Deep Neural Networks

Education facts

Code: MMA440

The goal of this course is to give an overview of the mathematical structures that appear in modern deep neural networks and enable students to understand current research literature in the field. To this end, the course starts with a concise introduction to the field of deep learning followed by deeper dives into the topics of equivariant neural networks, large width neural networks and geometrical aspects of explainable AI. The lectures are accompanied by computer labs where students learn the basics of implementing neural networks.

Entry requirements

General entry requirements and the equivalent of the courses MSG110 Probability Theory, MVG301 Programming with Python, and MMG500 Algebraic Structures. In addition to these requirements, it is also desirable with knowledge corresponding to the courses DIT013 Imperative Programming with Basic Object Orientation, MMA211 Higher Differential Calculus, and MMA201 Representation Theory.

Education offerings

Each offering has its own dates and conditions. Closed offerings are retained as history and do not mean that a new application is open.

• Mathematical Structures of Deep Neural Networks

  University of Gothenburg

  GÖTEBORG

  Start date: 17 January 2028

  End date: 22 March 2028

  Pace of study: 50 %