Machine Learning
2016/17

Instructors:

Alessandro Verri (alessandro.verri [at] unige.it)
Lorenzo Rosasco (lorenzo.rosasco [at] unige.it)

TA:

Raffaello Camoriano (raffaello.camoriano [at] iit.it)
Giulia Pasquale (giulia.pasquale [at] iit.it)

Class Times:

Mon: 11-13am; Tue: 2-4pm; Wed: 9-11am; Exceptions: See syllabus

Location:

DIBRIS-room 216; DIBRIS-lab Software 1 (SW1), 2nd floor

Office Hours:

Appointment by email

Course description

Machine Learning is a key to develop intelligent systems and analyze data in science and engineering. Machine learning engines enable intelligent technologies such as Siri, Kinect or Google self driving car, to name a few. At the same time machine learning methods help unlocking the information in our DNA and make sense of the flood of information gathered on the web, forming the basis of a new Science of Data. This course provides an introduction to the fundamental methods at the core of modern machine learning. It covers theoretical foundations as well as essential algorithms for supervised and unsupervised learning. Classes on theoretical and algorithmic aspects are complemented by practical lab sessions.

More info: http://computerscience.dibris.unige.it/course/ml/

2015/16 edition (ISML Mod. 2): /courses/isml2/isml2-2016/index.html

Prerequisites

The mathematical tools needed for the course will be covered in class,and include material form the following courses: Geometria (25882); Elementi di Statistica e Probabilità (67083), Calcolo Numerico (61804), Probabilità1 (52205).

Grading

Requirements for grading (other than attending lectures) are: attendance to classes + labs, project +discussion.

Syllabus

Follow the link for each class to find a detailed description, suggested readings, and class slides. Some of the later classes may be subject to reordering or rescheduling.

Date

Title

Lecturer

Resources

Class 1

Wed 28 Sep (9am - 10am)

Introduction

AV

Class 2

Mon 3 Oct

Statistical Learning Theory

AV

Class 28

Tue 4 Oct (2pm - 4pm)

Math Camp (2pm - 4pm; room 216)

 AV, RC, GP

Class 3

Tue 4 Oct (4pm - 6pm)

Lab (4pm - 6pm; Lab SW2) Octave/Matlab goal: basic matlab/octave+ data generation

AV, RC, GP

Class 4

Wed 5 Oct

Local Methods

AV

Class 29

Fri 7 Oct (2pm - 4pm)

Probability Camp (room 216)

GP

Class 5

Mon 10 Oct

Bias Variance Trade-Off

AV

Class 6

Tue 11 Oct

Lab on LM: K-NN, PW for classification

AV, RC, GP

Class 7

Mon 17 Oct

Least Squares Regression

AV

Class 8

Tue 18 Oct

Least Squares Classification

AV

Class 9

Wed 19 Oct (8:30am - 10am)

Lab LS/LDA

AV, RC, GP

C8 + C10 ING

Thu 20 Oct (11am - 1pm)

Class only for Engineering Students: Least Squares Classification + Feature Maps

AV

Class 10

Fri 21 Oct (2pm - 4pm)

Feature Maps

 AV

Class 11

Mon 24 Oct

Kernels

AV

Class 12

Tue 25 Oct

Lab Kernels

AV, RC, GP

Class 13

Wed 26 Oct

Regularization Networks and Representer Theorem

AV

Mon 31 Oct

No Class

Tue 1 Nov

 No Class

Class 14

Wed 2 Nov

Logistic Regression & Support Vector Machines

AV

Class 15

 Mon 7 Nov

Lab Loss functions

LR, RC, GP

Class 16+17

Tue 8 Nov

(TENTATIVE) Double Lab Learning Pipeline - time is (2pm - 6pm)

LR, RC, GP

Class 18

Wed 9 Nov

Dimensionality Reduction

LR

Class 19

Mon 14 Nov

Variable Selection & Sparsity

LR

Class 20

Tue 15 Nov

Lab Dimensionality Reduction and Variable Selection

LR, RC, GP

Class 21

Wed 16 Nov

Neural Networks

LR

Class 22

Mon 21 Nov

Multitask Learning

LR

Class 23

 Tue 22 Nov

Semisupervised Learning

LR

Class 24

Wed 23 Nov

Clustering & K-Means

LR

Class 25

Tue 29 Nov

Dictionary Learning

LR

Class 26

Wed 30 Nov

Density and support estimation

 LR

Class 27

19, 20, 21 Dec

Final Project Presentations

 LR, RC, GP

Class 30

TBD

Singular Value Decomposition - SVD Camp (room 216)

 RC

References

Further readings

Useful Links

Materials

Template of the Report