Program
1
Introduction
Short course overview, requirements. The principles of machine learning – how it works, what it can and cannot do. Basic concepts of mathematical statistics.
2
Unsupervised learning
Clustering - automatic search for similarity "templates" in the data. Reduction of data space, separation of key data features.
Dimension reduction - the separation of key data features.
3
Supervised learning
Forecasting different types of data, regression, and classification. Classification of textual data, metrics.
4
Deep learning
Acquaintance with neural networks - their types, principles of operation. Classification of visual data. Reuse of trained models.
Transfer learning.
5
Final project
Course completion project - The course completion project is an entirely independently completed practical task. Depending on the task, it is possible to choose several different algorithms, train them, choose the best one and think about the possibilities of their improvement.