The Applied Artificial Intelligence course prepares students to build real Artificial Intelligence solutions. This course explains the basics of Artificial Intelligence (machine learning, computer vision, natural language processing) from an extremely practical point of view. During the whole course, concepts will be explained and applied contextually. At the beginning of each module, a real-life problem will be presented to students. All the explained AI concepts that will be useful to solve the relative business problem.

Audience

Undergraduate senior students from IT related academic programs, for example, computer science, software engineering, information systems etc.

Objectives

After completing this course, students will:

• Have a solid understanding of what AI really is and why it is considered as a crucial asset for many Businesses.
• List AI use cases for different industries.
• Explain Deep Learning and its most important applications.
• Explain Machine Learning and its most common applications.
• Describe examples of unsupervised and supervised learning.
• Describe the software packages that can be used to train AI algorithms.

 Prerequisites Skills 

• Computer science fundamentals
• Basic knowledge of applied math, algorithms, and data modelling
• Basic knowledge of probability and statistics
• Basic knowledge of a programming language

Duration

24 hours (base module)

40 Hours (complete module)

Module 1 – Artificial Intelligence

• What Artificial Intelligence really is
• What can be actually done with it.
• Which are the most popular AI use cases.
• The fundamental notions (Neural Networks, Supervised and unsupervised Learning).
• The packages needed to manipulate data, train algorithms and perform predictions.

Module 2 – Machine Learning aided Fraud Detection

• This module explains how to use Machine Learning algorithms to hinder fraudulent activity on electronic payments. The objective is reached by learning from data how fraudsters behave and by spotting anomalous transactions using unsupervised algorithms. All concepts will be explained during a one day practical project.

Module 3 – Artificial Intelligence-empowered Cyber Security

• This module shows how Artificial Intelligence can help security analysts to face the most difficult problems: 0-day attacks, persistent threats and organized threats (botnets, etc.). The objective is reached by identifying anomalies in the traffics data and modelling the whole network as a graph. All concepts will be explained during a one day practical project.

Module 4 – Computer Vision

• What computer vision is.
• The most common use cases.
• The necessary data pre-processing activities (resizing, rotating, transforming colours, etc.).
• How features are extracted from images by Deep Neural Network architectures.
• The details of the convolutional network architecture
• How classification is performed over image data.
• All concepts will be explained during a one day practical project

Module 5 – Natural Language Processing

• What computer vision is.
• The most common use cases.
• The (almost often) necessary data pre-processing steps (text cleaning, stopwords removal, lemmatization, etc.).
• The most common encoding techniques (TF-IDF, word embedding, text embedding, BERT).
• The most popular supervised algorothms for text dta (Naïve Bayes, RNN, etc.).
• All concepts will be explained during a one day practical project

Module 6 – Putting Artificial Intelligence in production

• What Big Data means and why it is si important for deploying Artificial Intelligence solutions.
• How to use trained artificial intelligence models on real production environments.
• How to perform real time (or near real time) scoring.
• Which are the most common architectures.
• The most common pitfalls and how to avoid them.