Machine Learning: Theory and Algorithms (MALTA), 2020-2021
The course introduces the mathematical foundations of machine learning.
Its first goal is to formalize the main questions behind machine learning: What is learning? How can a machine learn? Is learning always possible? How do we quantify the resources needed to learn?
To this purpose, the course presents the probably-approximately correct (PAC) learning paradigm.
Its second goal is to present several key machine learning algorithms and show how they follow from general machine learning principles.
The course has a theoretical focus, and the student is assumed to be comfortable with basic notions of probability, linear algebra, analysis, and algorithms.
Teacher: Giovanni Neglia.
Main reference:
Shai Shalev-Shwartz and Shai Ben-David, Understanding Machine Learning: From Theory to Algorithms, available here.
There are also video lessons from Shai Ben-David, and lecture notes from Shai Shalev-Shwartz (together with videos in Hebrew if you prefer it to English).
Evaluation
30% classwork (a 10-minute test at every lesson, only 5 best marks will be considered), 30% a mid-course home assignement, 40% final exam.
Lessons
Lessons will be from 14.00 to 17.15 in Lucioles campus.
For each lesson, the corresponding sections of the book and the corresponding videos of Shai Ben-David's course are listed.
First lesson (September 15, room A2): chapters 1-2, videos for lectures 1-2.
Second lesson (September 29, room A1): chapters 3-4, videos for lectures 3-5.
Third lesson (October 6, room A1): chapter 5, section 6.1 and the beginning of section 6.2, videos for lecture 6 (you can skip from minute 53 until 1:15), lecture 7 (watch until minute 50) and lecture 8 (watch from minute 11 until minute 46).
Fourth lesson (October 13, room A1): sections 6.2-6.4, sections 9.1.1 and 9.1.3, videos for lectures 7, 8, and 9 (note that the videos do not cover all the sections).
Fifth lesson (October 20, room A1): chapter 10 (without proofs), section 12.1.1, videos for lectures 17 (from minute 45), 18, 19, and 20 (up to 1:04:00).
Sixth lesson (October 27, room A2): beginning of section 12.2, sections 12.3, 9.2, 9.3, videos for lectures 20 (from 1:04:00), 21, and 22 (you can skip between 0:14:30 and 0:48:20).
Seventh lesson (online): sections 7.1-7.3 (without proofs), chapter 11, videos for lectures 12 (from 0:44:40), 13, 14, and 15 (up to 0:35:00). Note that the videos do not cover sections 11.2 and 11.3.
Assignment
Assignment text. Hand your solution to the teacher before the start of the lecture on October 20th. If you cannot attend the lecture, send the solution by email.
Exam
The exam will be on November 10th in room E+142, Templiers, from 14.00 to 17.00.
