Course 2017-2018 a.y.

Classes: 25 (II sem.)

Instructors:
Class 25: RICCARDO ZECCHINA

The goal of this course is to provide the students with the conceptual, mathematical and algorithmic tools that are needed for a rigorous and modern understanding of machine learning. A central topic is the connection with Bayesian inference and, more in general, the critical discussion on how and to which extent machine learning methods become essential in large scale data analysis and in artificial intelligence.
The students implement data analysis algorithms and learn to use the modern scientific libraries; the course includes an original algorithmic development project.

The final mark is the sum of partial marks, obtained through

• 2 partial written exams.

or

• a general written exam.
  

Each partial exam assigns 16 points; the general exam assigns 32 points.
30 cum laude is obtained with 31 points or more.

The exam takeS place in the informatics classroom.

• D. MAC KAY, Information Theory, Inference, and Learning Algorithms, Cambridge University Press, 2003.
• I. GOODFELLOW, Y. BENGIO, A. COURVILLE,Deep Learning, MIT Press, 2016.
• Handouts.

Intermediate level programming, familiarity with the Python numerical and scientific library stack, fundamentals of analysis, fundamentals of linear algebra, fundamental of statistics, fundamentals of optimization theory.