Machine Learning Basics for Information and Communications Engineering

Course Overview

Intended learning outcomes

At the end of the course, the student will be able to:

• distinguish between different types of machine learning problems
• know which technique to use to solve them and how to implement it
• understand the mathematical fundamentals of artificial neural networks
• use them into a deep learning (DL) framework
• apply the acquired knowledge to solve information and communication engineering related challenges
• know the latest DL solutions in the domain of communications

Teaching methodology

The course will cover important theoretical aspects in details that are typically behind the functioning of most artificial intelligence systems and will use Python to implement the studied algorithms.

Course content

Topics covered in the course are the following:

• Introduction to the course
  • What is machine learning for information and communication engineering (ICE), current applications
• Fundamentals of machine learning, from problem analysis/formulation to its solution and evaluation
  • Type of data, learning problems, learning techniques and evaluation methods
• Neural networks
  • Artificial neural networks, back-propagation, gradient descent, activation functions
• Introduction to deep learning for ICE
  • Relevant network architectures, e.g., CNNs, RNNs, LSTM, Transformers, Diffusion Models
• Deep learning applications to ICE
  • Learning to decode
  • Autoencoders and their application in communications
  • Generative adversarial networks (GANs) and their application in communications

Prior knowledge expected

Basics of linear algebra, statistics and programming.