PhD-Summer Course- June 30th- July 4th,  Genova,  Italia


Regularization Methods for Machine Learning - RegML 2014
Instructors: Francesca Odone, Lorenzo Rosasco

A 20 hours advanced machine learning course  including theory classes and practical laboratory session. The course covers  foundations as well as  recent advances in Machine Learning with  emphasis on  high dimensional data and  a core set techniques, namely regularization methods. In many respect the course is compressed version of the 9.520 course at MIT

The course started in 2008 has seen an increasing national and international attendance  over the years with a peak of  85 participants in 2013.

Registration CLOSED, COURSE ACTIVATION CONFIRMED! We will not accept further registrations since we have reached  a maximum number. If you have registered, you will  soon receive a  confirmation e-mail.

See here list of participants, let us know if your information are not correct.

NOTE: the course has no registration fee but participants need to take care of their accommodations -- see below for a list of hotels.

Basic Info   |    Synopsis   |    Syllabus   |    More Info

In collaboration with:


Basic Info

Venue:  Classes will take place at the Department of Informatics Bioengineering Robotics and Systems Engineering (DIBRIS) of the University of Genova in Via Dodecaneso 35, 16146 Genova. See here for directions and travelling information

Genova is in the region of Liguria in the Italian Riviera (see here or here for some nice pics and a video)  

Accomodations: Here you can find a list of hotels near the department (~ 20' walk) or in the city centre (~20' by bus).

Lunch: Here is a list of places where you can go for lunch. And here is a link to the online google map


Synopsis

Understanding how intelligence works and how it can be emulated in machines is an age old dream and arguably one of the biggest challenges in modern science. Learning, with its principles and computational implementations, is at the very core of this endeavor. Recently, for the first time, we have been able to develop artificial intelligence systems able to solve complex tasks considered out of reach for decades. Modern cameras recognize faces, and smart phones voice commands, cars can see and detect pedestrians and ATM machines automatically read checks. In most cases at the root of these success stories there are machine learning algorithms, that is softwares that are trained rather than programmed to solve a task. Among the variety of approaches to modern computational learning, we focus on regularization techniques, that  are key to high- dimensional learning. Regularization methods allow to treat in a unified way a huge class of diverse approaches, while providing tools to design new ones. Starting from classical notions of smoothness, shrinkage and margin, the course will cover state of the art techniques based on the concepts of geometry (aka manifold learning), sparsity and a variety of algorithms for supervised learning, feature selection, structured prediction, multitask learning and model selection. Practical applications for high dimensional problems, in particular in computational vision, will be discussed. The classes will focus on algorithmic and methodological aspects, while trying to give an idea of the underlying theoretical underpinnings. Practical laboratory sessions will give the opportunity to have hands on experience.
 


Syllabus & Schedule

- each class is 90 min. no breaks -

Day Time Class Title

Mon

9:30am-11am 1

Welcome+ Introduction to Learning

Mon

11:30pm:1pm 2 Kernels, Dictionaries, and Regularization

Mon

2:30pm:4 pm 3 Regularization Networks and Support Vector Machines
Tue
9:30am-11am 4 Spectral methods for supervised learning
Tue 11:30am-1pm 5 Manifold regularization
Tue 2:30pm:4pm 6 Lab 1 - Binary classification and model selection
Wed
9:30am-11am 7 Error Analysis and Parameter Choice
Wed 11:30am-1pm 8 Lab 2 - Spectral filters and multi-class classification
Wed Afternoon - FREE
Thur
9:30am-11am 9 Sparsity-based Regularization
Thur 11:30am-1pm 10
Structure Sparsity and Multiple Kernel Learning
Thur 2:30pm:4pm 11
Lab 3 - Sparsity-based learning
Fri 9:30am-11am 12
Dictionary Learning
Fri 11:30am-1pm 13
Applications to high dimensional problems


More Info

Credits and Exams: If you attend most of the classes you will be attributed 2 credits (according to the ECTS grading scale). The credits attribution will be reported on the certificate of attendance we will handle at the end of the course.


If you need an evaluation the exam will consist in a brief report (~ 5 pages + 1 page of figures) of the labs.
Submission deadlines: TBA. Submit your report (one or multiple authors are fine) by sending an email to both Francesca and Lorenzo and specifying the type of evaluation you need (eg., passed / ranking / marking...)

Prerequisites: Basic Multivariate Calculus, Basic Probability Theory, Matlab.