RegML 2017 @ SIMULA-Oslo
Regularization Methods for Machine Learning

Course at a Glance

This year (2-6 May 2017) RegML will be hosted and organized by SIMULA in Oslo

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.


RegML is a 22 hours advanced machine learning course including theory classes and practical laboratory sessions. 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 over 90 participants in 2014.


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


Notification of acceptance: To be announced.

Related courses:

Basic Info

Venue

School will take place at Simula Research Laboratory, Martin Linges vei 25, 1364 Fornebu, Norway. Consult our page on directions and travelling information for more details on how to get to Simula. All lectures will take place in “Klasserommet” auditorium at Simula. All project work will take place in assigned workspaces at Simula (detailed later).


Accommodations

There are several hotels and options for accommodation available near Simula, and the city center is 25 minutes away by bus.
Here you can find a list of hotels near Simula.


Lunch

Sandwiches will be provided at 12:30. Hot meal and other options can be found in the canteen, just outside Simula.


Instructors

Lorenzo Rosasco

Università di Genova
Istituto Italiano di Tecnologia
Massachusetts Institute of Technology

lorenzo (dot) rosasco (at) unige (dot) it

Valeriya Naumova

Simula Research Laboratory

valeriya (at) simula (dot) no

Teaching Assistants

Luigi Carratino

Università di Genova

luigi (dot) carratino (at) dibris (dot) unige (dot) it

Guillaume Garrigos

Istituto Italiano di Tecnologia

guillaume (dot) garrigos (at) iit (dot) it

Zeljko Kereta

Simula Research Laboratory

zeljko (at) simula (dot) no

Timo Klock

Simula Research Laboratory

timo (at) simula (dot) no



Workshop

Invited Speakers

Gilles Blanchard

University of Potsdam

Arthur Gretton

University College London

Leonidas Lefakis

Zalando Research

Miguel Rodriguez

University College London

Alessandro Rudi

Istituto Italiano di Tecnologia

Dino Sejdinovic

University of Oxford

Syllabus

CLASS DAY TIME SUBJECT FILES
1Tue 5/29:00 - 10:30Introduction to Machine Learning
2Tue 5/211:00 - 12:30Local Methods and Model Selection
3Tue 5/214:00 - 16:00Laboratory 1: Local Methods for Classification
4Wed 5/313:00 - 14:30Tikhonov Regularization and Kernels
5Wed 5/315:00 - 17:00Laboratory 2: Binary classification and model selection
6Thu 5/49:00 - 10:30Early Stopping and Spectral Regularization
7Thu 5/411:00 - 12:30Regularization for Multi-task Learning
8Thu 5/414:00 - 16:00Laboratory 3: Spectral filters and multi-class classification
9Fri 5/59:00 - 10:30Sparsity Based Regularization
10Fri 5/511:00 - 12:30Structured Sparsity
11Fri 5/5 14:00 - 16:00Laboratory 4: Sparsity-based learning
-Sat 5/6 Workshop

The deadline for applications is April 1

Important: Applicants who are not from SIMULA must also send their updated CV or resume to regml2017applications@gmail.com

Organizers

Valeriya Naumova

Simula Research Laboratory

valeriya (at) simula (dot) no

Lorenzo Rosasco

Università di Genova
(also Istituto Italiano di Tecnologia and Massachusetts Institute of Technology)

lorenzo (dot) rosasco (at) unige (dot) it

Luigi Carratino

Università di Genova

luigi (dot) carratino (at) dibris (dot) unige (dot) it

Guillaume Garrigos

Istituto Italiano di Tecnologia

guillaume (dot) garrigos (at) iit (dot) it