LEARNING OUTCOMES

This course provides an introduction to machine learning and statistical pattern recognition. Topics include: (1) Pattern recognition basics and theory. (2) Supervised learning ideas and methods. (3) Unsupervised learning ideas and some relevant methods. (4) Machine learning workflows and best practices. The course will also cover relevant success stories, and possible applications and case studies in the fields of robotics and smart industrial automation

AIMS AND LEARNING OUTCOMES

After successfully attending the course, the student will be able to:

• demonstrate knowledge of a range of techniques and problems in machine learning and pattern recognition, including the underlying scientific and technical rationale
• apply selected techniques to relevant problems
• code simple and medium-complexity machine learning methods using standard programming tools, without being limited to using software libraries
• tackle the workflow of a machine learning assignment from data wrangling to result presentation
• use critical thinking to analyse a problem and select the appropriate machine learning method to apply

PREREQUISITES

• Basic knowledge of calculus, linear algebra, geometry, which are typically acquired during the first or second year in any engineering curriculum
• Basic, but operational, knowledge of Matlab or Python programming

More basic topics (elements of probability, of statistics, of optimisation) will be covered during the course

TEACHING METHODS

Lectures, guided labs, homework assignments

SYLLABUS/CONTENT

• Introduction, basic concepts, types of problems
• Linear thershold classifiers
• Probabilities; Bayesian decision theory (the Naive bayes classifier)
• Linear regression as a simple learning problem
• Optimisation (convexity, criteria, gradient descent, stochastic methods)
• Statistics and learning (sampling, parameter estimation)
• Parametric and non-parametric methods (Gaussian mixtures, nearest neighbour rules, decision trees/forests)
• Evaluation of classifiers (methodology, quality indices)
• Neural networks (history, perceptrons, multilayer perceptrons, the error back-propagation algorithm, deep learning)
• Unsupervised learning (clustering methods)
• Mapping and input space transformations (PCA, nonlinear embedding methods, kernel methods, support vector machines)

RECOMMENDED READING/BIBLIOGRAPHY

Course slides/handouts

For a detailed bibliograpy please refer to the course Aulaweb page (from https://corsi.unige.it/9269#chapter-5 open Manifesto degli Studi, look up Machine Learning and click it)