93319 - INTRODUZIONE ALL'APPRENDIMENTO AUTOMATICO

Learning outcomes

During the course, the student will be introduced to the complex themes pertaining to the simulation of intelligent behavior by means of machines, with practical experimentation of basic machine learning techniques for different tasks: supervised, unsupervised, with reinforcement. The course will also provide rudiments of image processing, since images will be extensively used as experimental test bench for the aforementioned learning techniques.

Course contents

The first part of the course provides a general introduction to the field of machine learning, in its typical forms: supervised, unsupervised, with reinforcement. Traditional topics such as decision tree learning, logistic regression, Bayesian networks and Support Vector Machines will be covered.

The second part of the course is focused on Neural Networks, and their typical learning mechanism: the backpropagation algortihm. We shall discuss the main types of neural nets: feed forward, convolutional, recurrent, and their practical applications. We shall also investigate techniques to visualize the effect of hidden units (tightly related to deep dreams and inceptionism) as well as several generative approaches comprising Generative Adversarial Networks. Thematics relative to Object Detection and Semantic Segmentation will be also briefly discussed too.

Readings/Bibliography

Teacher's slides.

During the course, additional links to relevant documents and sites will be provided.

Teaching methods

Frontal lessons integrated with practical exemplifications

We also foresee additional laboratories held by tutors for a total of 12 hours

Assessment methods

Individual project on a topic defined by the teacher, possibly integrated by a written quiz.

Teaching tools

The course will make use of several opens source libraries for Machine Learning. In particular we shall mostly use

• scikit-learn
• Tensorflow/Keras

Office hours

See the website of Andrea Asperti