Our journey starts from the title of this course.
Artificial Intelligence (AI), Machine Learning (ML) and Pattern Recognition (PR) refer, to a certain degree, to the same, underlying concepts and are often used interchangeably. Defining AI, ML and PR requires more than a lesson and is out of our scope, however, in the following, we provide simple and non-exhaustive definitions of these three notions, in order to get the general idea of what we are going to study.
Artificial intelligence has nowadays become the term reserved to general public to define any applications that rely on computer-automatized learning routines. Since a computer, by definition, automatizes processes, the “learning” term distinguishes AI-powered computer applications from the others computing routines. Note also that the term computer is not related to electronics/circuitry only, and it encompasses the biological counterparts too.
Machine learning is used in tech-based contexts, to define the methodology of AI research. In ML, we make use of data collected in the wild to train algorithms (models). The aim of the training consists of learning the underlying structure contained in the data. The aim of this course is to explain the motivation and application of some ML models and their training procedures.
Finally, pattern recognition is somewhat between the other two terms described above and defines the aim of ML. The underlying structure contained in the data is precisely called pattern and is what we are interested to learn automatically. Nowadays, it is less used than the other two terms.
A short summary
The following widget is a timeline of important events in Artificial Intelligence. Note that the timeline contains YouTube resources so, if you use it, you silently accept YouTube cookies