Mathematical Structures of Deep Neural Networks
About
The course introduces the basic concepts of machine learning and deep learning. You will learn to use the most important deep learning architectures and training schemes. You will also learn to train a neural network using common deep learning frameworks in Python.
The course content covers a wide range of topics in machine learning, including types of learning problems, bias-variance tradeoff, overfitting, regularization, feed-forward neural networks, backpropagation, gradient-based optimizer, Glorot- and He initialization and regularizing neural networks.
The course is given every other year, in the spring semester in even-numbered years.
Prerequisites and selection
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.
Selection
All eligible applicants who have applied before the deadline will be granted a place.
Facilities
Mathematical Sciences is a joint department of Chalmers/University of Gothenburg. Your education takes place in the spacious and bright premises of Mathematical Sciences at the Chalmers campus Johanneberg, where there are lecture halls, computer rooms and group rooms. Here you can also find student lunch room and reading room, as well as student counsellors and student office.
