Course description
Recent development in machine learning have led to a surge of interest in artificial neural networks (ANN). New efficient algorithms and increasingly powerful hardware has made it possible to create very complex and high-performing ANNs. The process of training such complex networks has become known as deep learning and the complex networks are typically called deep neural networks.
The aim of this course is to introduce students to common deep learnings architectues such as multi-layer perceptrons, convolutional neural networks and recurrent models such as the LSTM. Basic concepts in machine learning till also be introduced. The course consists of a series of lectures and computer exercises. The programming environment will be python (Jupyter notebook) together with the deep learning libraries Keras and Tensorflow.
The course will be given in flipped classroom mode, with students watching recorded lectures together with online quiz meetings with discussions.
Pre-requisites/requirements
- Programming: Basic knowledge.
- Mathematics: Calculus in one and several variables and linear algebra
- Standard desktop/laptop computer and internet connection
Schedule
- 27 Sept 10:15-12:00 Introduction to ML and DL
- 1 Oct 14:15-15:00 The MLP-1
- 4 Oct 10:15-12:00 The MLP-2
- 7 Oct 14:15-15:00 CNN, part 1
- 8 Oct 14:15-15:00 CNN, part 2
- 11 Oct 10:15-12:00 Autoencoder and GAN
- 14 Oct 14:15-15:00 Recurrent networks
- 22 Oct 13:15-17:00 Presentations of project work
- 25 Oct 9:15-17:00 Presentations of project work
All teaching events will be online.
Personnel
Course organiser: Mattias Ohlsson
Teachers: Mattias Ohlsson and Patrik Edén
Examination
Written report and oral presentation of a deep learning computer project
Registration
Registration closes on the 19th September 2021 (hard deadline).
Registration is now closed.