Introduction to Machine Learning and Deep Learning Using Python and PyTorch

Course description

The course introduces students to machine learning (ML) and deep learning (DL), starting from fundemental principles and a quick refreshing of probabilities and statistics. It presents concepts such as supervised learning and unsupervised learning, presents concrete cases such as classification, regression, and clustering. It then continues to DL and presents artificial neural networks, common architectures and progresses to state-of-the-art models that are used currently both in industry and research.

Emphasis is givven both:

• on the theoritical aspect, by introducing each concept thoroughly with clear mathematical notation
• on the practical aspect: the students get to setup a workign environment from the first lecture and are given examples on every lecture in order to see how each new concept is applied in practice

The course also introduces students to other practical matters of ML and DL development such as managing models, viewing real-time plots when training using well-established tools, managing different projects and configurations, and more.

Prerequisites and Selection

Prerequisite courses, or equivalent

Basic understanding of probability theory:

• definition of probability
• conditional probability
• probability density function
• random variable

Average skill in programming in languages such as Python (preferred), MATLAB, R, or similar, which includes:

• arrays and matrices manipulation
• functions
• classes (and basic object-oriented-programming concepts)
• libraries/modules: how to import and use them (on a basic level)

Examples of courses that provide sufficient knowledge in programming (each one individually is enough, not all are required):

• H7F6034: Introduction to Programming using Python
• H7F5300: Get started with R – Programming Basics, Data Analysis and Visualisation
• H7F6003: Intermediate R –Data Science and Visualization Techniques Beyond Base R

Selection

Selection will be based on:  

1. the relevance of the course syllabus for the applicant’s individual study plan/research (according to written motivation). 
2. start date of doctoral studies (priority given to earlier start date). 

Course director

Vasileios Papapanagiotou

Course syllabus

H7F6081

Department

Department of Medicine, Huddinge

Doctoral programme

**Not within a doctoral programme

Type of course

Statistics Software

Keywords

artificial intelligence, machine learning, deep learning