Registration site open: VISION AND SPORTS SUMMER SCHOOL

Prague, 27-31 August 2012

http://cmp.felk.cvut.cz/summerschool2012/
email: vs3@cmp.felk.cvut.cz

application deadline: 15 May 2012

Organized in collaboration with NIFTi
http://www.nifti.eu

OVERVIEW

Vision and Sports is a special special kind of summer school. In
addition to a broad-range of lectures on state-of-the-art Computer
Vision techniques, it offers exciting sport activities, such as
Tennis, Archery, Kung-Fu and Ultimate Freesbee. Sports are organized
by the same internationally renowned experts who deliver the lectures.
The school offers the best of both worlds to participants:
high-quality teaching on Computer Vision, and lots of fun with a
variety of attractive sports. This offers plenty of opportunity for
personal contact between students and teachers.

The Vision and Sports Summer School covers a broad range of subjects,
reflecting the diversity of Computer Vision. Each lecture will cover
both basic aspects and state-of-the-art research. Every day there are
two Computer Vision classes and one sports session. The classes
include both lectures and practical exercises.

The school is open to about 60 participants, and is targeted mainly to
young researchers (Master students and PhD students in particular).

Following the 5-day summer school, there will be a robotic workshop
organized by the EU project NIFTi (www.nifti.eu). The participants
will apply vision algorithms on the NIFTi mobile robot(s) and move the
robot around using the Ladybug3 camera the robot is equipped with.
Attendance to this workshop is optional and limited to a small number
of attendees. http://cmp.felk.cvut.cz/niftivisionworkshop2012/

TEACHERS

Jiri Matas
Czech Technical University

Phil Torr
Oxford Brookes University

Carsten Rother
Microsoft Research Cambridge

Christoph Lampert
IST Austria

Patrick Perez
Technicolor Research and Innovation

Silvio Savarese
University of Michigan

Vittorio Ferrari
University of Edinburgh

Bastian Leibe
RWTH Aachen University

Ondrej Chum
Czech Technical University

COMPUTER VISION LECTURES

Current list of topics:

Local feature extraction
Large-scale specific object recognition
Single-view and multi-view object categorization
MRF/CRF in Computer Vision
Semantic segmentation
Weakly supervised learning of visual models
Kernel Methods in Computer Vision
Tracking in video

SPORT ACTIVITIES

Tennis, Volleyball, Unihockey, Archery, Table Tennis, Soccer, Archery,
Ultimate Freesbee, Kung-Fu, Basketball, Tai chi, Badminton

APPLICATION

The school is open to about 60 participants. Please apply online at

http://cmp.felk.cvut.cz/summerschool2012/

Although priority will be given to young researchers (Master/PhD
students in particular), applications from senior researchers and
industrial professionals are welcome as well. The registration fee for
PhD students is only 300 Euro. This fee includes all classes, sports
activities, coffee breaks, lunches, and a social dinner. For hotel
accommodation, students will get discount rates on hotels affiliated
with the school.

Applicants should apply before 15 May 2012.
Notification of acceptance will be sent by 31 May 2012.

MORE INFORMATION

http://cmp.felk.cvut.cz/summerschool2012/

2nd Call For Papers – Special Issue on Learning Semantics in Machine Learning

**Overview**
A key ambition of AI is to render computers able to evolve and interact
with the real world. This can be made possible only if the machine is able
to produce an interpretation of its available modalities (image, audio,
text, etc.) which can be used to support reasoning and taking appropriate
actions. Computational linguists use the term “semantics” to refer to the
possible interpretations of natural language expressions and there is
recent work in “learning semantics” – finding (in an automated way) these
interpretations. However, “semantics” are not restricted to the natural
language (and speech) modality, and are also pertinent to visual
modalities. Hence, knowing visual concepts and common relationships
between them would certainly provide a leap forward in scene analysis and
in image parsing akin to the improvement that language phrase
interpretations would bring to data mining, information extraction or
automatic translation, to name a few.

Progress in learning semantics has been slow mainly because this involves
sophisticated models which are hard to train, especially since they seem
to require large quantities of precisely annotated training data. However,
recent advances in learning with weak, limited and indirect supervision
led to the emergence of a new body of research in semantics based on
multi-task/transfer learning, on learning with semi/ambiguous/indirect
supervision or even with no supervision at all. Hence, this special issue
invites paper submissions on recent work for learning semantics of natural
language, vision, speech, etc.

Papers should address at least some of the following questions:
– How should meaning representations be structured to be easily
interpretable by a computer and still express rich and complex knowledge?
– What is a realistic supervision setting for learning semantics?
– How can we learn sophisticated representations with limited supervision?
– How can we jointly infer semantics from several modalities?

**Dates**
Submission deadline: May 1, 2012
First review results: July 30, 2012
Final drafts: September 30, 2012

**Submissions**
Papers must be submitted online, selecting the article type that indicates
this special issue. Peer reviews will follow the standard Machine Learning
journal review process. It is the policy of the Machine Learning journal
that no submission, or substantially overlapping submission, be published
or be under review at another journal or conference at any time during the
review process. Papers extending previously published conference papers
are acceptable, as long as the journal submission provides a significant
contribution beyond the conference paper, and the overlap is described
clearly at the beginning of the journal submission. Complete manuscripts
of full length are expected, following the MLJ guidelines in
http://www.springer.com/computer/ai/journal/10994 .

**Guest Editors**
Antoine Bordes (antoine.bordes@utc.fr)
Léon Bottou (leon@bottou.org)
Ronan Collobert (ronan@collobert.com)
Dan Roth (danr@illinois.edu)
Jason Weston (jweston@google.com)
Luke Zettlemoyer (lsz@cs.washington.edu)

CALL FOR CONTRIBUTIONS: Object, functional and structured data : towards next generation kernel-based methods

ICML 2012 Workshop, June 30, 2012, Edinburgh, UK.

https://sites.google.com/site/nextgenkernels/
========================================================================
Important dates
Submission due by May 7, 2012.
Author Notification, May 21, 2012.
Workshop, June 30, 2012.

Topic
This workshop concerns analysis and prediction of complex data such as objects, functions and structures. It aims to discuss various ways to extend machine learning and statistical inference to these data and especially to complex outputs prediction. A special attention will be paid to operator-valued kernels and tools for prediction in infinite dimensional space.

Context and motivation
Complex data occur in many fields such as bioinformatics, information retrieval, speech recognition, image reconstruction, econometrics, biomedical engineering. In this workshop, we will consider two kinds of data: functional data and object or structured data. Functional data refers to data collected under the form of sampled curves or surfaces (longitudinal studies, time series, images). Analysis of these data as samples of random functions rather that a collection of individual observations is called Functional Data Analysis (FDA). FDA involves statistics in infinite-dimensional spaces and is closely associated to operatorial statistics. Its main approaches include functional principal component analysis and functional regression. Many theoretical challenges remain open in FDA and attract an increasing number of researchers.
Object and structure data exhibit an explicit structure like trees, graphs or sequences. For instance, documents, molecules, social networks and again images can be easily encoded as objet structured data. For the two last decades, both machine learning and statistics communities have developed various approaches such as graphical probabilistic models as well as kernel methods to take into account the structure of the data. In the meantime, FDA has been extended to Object Data Analysis which deals with samples of object data.
However, most of the efforts have been concentrated so far on dealing with complex inputs. In this workshop, we would like to emphasize the problem of complex outputs prediction which is involved for instance in multi-task learning, structured classification and regression, and network inference. All these tasks share a common feature: they can be viewed as approximation of vector-valued functions instead of scalar-valued functions and in the most general case, the output space is an Hilbert space. A promising direction first developed in (Micchelli and Pontil, 2005) consists in working with Reproducing Kernel Hilbert Spaces with operator-valued kernels in order to get an appropriate framework for regularization. There is thus a strong link between recent works in machine learning about prediction of multiple or complex outputs and functional and operatorial statistics.
This workshop aims at bringing together researchers from both communities to 1) provide an overview of existing concepts and methods, 2) identify theoretical challenges and (3) discuss practical applications and new tasks.

Invited speakers
Yasemin Altun (Google)
Frédéric Ferraty (University of Toulouse, France)
Arthur Gretton (Gatsby Unit, UCL MPI for Intelligent Systems, UK)
Neil Lawrence (University of Sheffield, UK)
Steve Marron (University of North Carolina, USA)
Charles Micchelli (University of Albany, USA)

Call for contributions
We invite short, high-quality submissions on the following topics:

* complex output learning
* structured output prediction
* functional data analysis
* object data analysis
* operator-valued kernels
* operator-based statistics
* joint-kernel maps
* statistical dynamics
* applications (non exhaustive list) : signal and image processing, bioinformatics, natural language processing, time series modeling …

Submission guidelines
Submissions should be written as extended abstracts, no longer than 4 pages in the ICML latex style. ICML style files and formatting instructions can be found at The submissions should include the authors’ name and affiliation since the review process will not be double blind. The extended abstract may be accompanied by an unlimited appendix and other supplementary material, with the understanding that anything beyond 4 pages may be ignored by the program committee. Please send your submission by email to nextgenkernelicml2012@gmail.com before May 7, 2102 at midnight PDT. Recently-published work is allowed.
We expect to select contributions for the spotlight and poster sessions. Authors will receive a notification by May 21, 2012.

Organizers
Florence d’Alché-Buc (University of Evry & INRIA-Saclay, France)
Hachem Kadri (INRIA-Lille, France)
Massimiliano Pontil (University College London, UK)
Alain Rakotomamonjy (University of Rouen, France)

Website admin: Céline Brouard (University of Evry, France)

Contact: nextgenkernelicml2012@gmail.com

EWRL 2012 – Final Call for Papers

Key Facts:

EWRL 2012: The 10th European Workshop on Reinforcement Learning
Location: Edinburgh, Scotland (2-days ICML workshop)

Dates:
EWRL 2012: June 30-July 1
Submission Deadline: April 15, 2012
Notification Date: May 15, 2012

Proceedings published in JMLR W&C, Vol. 24

URL: http://ewrl.wordpress.com/ewrl10-2012/

Organizers: Marc Deisenroth, Csaba Szepesvari, Jan Peters
***************************************************************

EWRL 2012 aims to serve as a forum to discuss the current
state-of-the-art and future research directions in the continuously
growing field of reinforcement learning. We intend to make this an
exciting event not only for the European RL community but also
international researchers from related areas with many opportunities to
share new knowledge and encourage collaborative work.

The main question of this workshop is to discuss, how other statistical
learning techniques may be used to developed new RL approaches in order
to achieve properties including higher numerical robustness, easier use
in terms of open parameters, probabilistic and Bayesian interpretations,
better scalability, the inclusions of prior knowledge, etc.

We are calling for papers from the entire reinforcement learning
spectrum, with the option of either 2 page short papers or longer 8 page
JMLR W&C Proceedings format research papers. We encourage a range of
submissions to encourage broad discussion. We will publish selected
papers in the prestigious JMLR W&C Proceedings, Vol. 24.

Double submissions are allowed but must be clearly indicated. However in
the event that an EWRL
paper is accepted to another conference proceedings or journal, it will
not be reprinted in the official EWRL proceedings. The paper would still
be considered, however, for acceptance and presentation at EWRL.
Double submissions must be clearly labelled as such (e.g., add a
footnote on the first page). In case your ICML submission exceeds EWRL’s
page limit, don’t worry too much about it: submit the ICML paper.

We encourage submissions from a range of sub-topics including
(but not limited to):

– Reinforcement Learning Theory
– Function Approximation in Reinforcement Learning
– Current Progress in Bandit Regret Bounds
– MDPs, POMDPs
– Exploration vs Exploration Tradeoff
– Multi-Agent Reinforcement Learning
– Policy Search
– Actor-Critic Methods
– Bayesian Control Approaches
– RL Benchmark Problems
– Real-world Applications
– Robot RL

Keynote Speakers:

Richard Sutton (University of Alberta)
Shie Mannor (Technion)
Martin Riedmiller (University of Freiburg)
Drew Bagnell (CMU) (tentative)

Submission deadline: April 15, 2012
Page limit: 2 pages for short papers and 8 pages for regular
papers.
Paper format: JMLR W&C, Vol. 24
Style file:
http://www.tex.ac.uk/tex-archive/help/Catalogue/entries/jmlr.html

For more information, see http://ewrl.wordpress.com/ewrl10-2012/

2012 IEEE International Workshop on Machine Learning for Signal Processing

****** Apologies for cross-posting **********

***************************************************************************************
2012 IEEE International Workshop on Machine Learning for Signal Processing
September 23-26, 2012.
Santander, Spain
http://mlsp2012.conwiz.dk
**************************************************************************************

The 22nd MLSP workshop in the series organized by the Signal Processing Society MLSP Technical Committee will present the most recent and exciting advances in data analysis for signal processing problems through tutorials, keynote talks, as well as special and regular single-track sessions.

* TOPICS OF INTEREST include but are not limited to:
– Learning theory and techniques
– Graphical models and kernel methods
– Data-driven adaptive systems and models
– Pattern recognition and classification
– Distributed, Bayesian, subspace/manifold, and sparsity-aware learning
– Multiset data analysis and multimodal data fusion
– Perceptual signal processing in audio, image and video
– Cognitive information processing
– Applications, including:
– Speech, audio and video
– Music processing
– Biomedical imaging
– Communications
– Bioinformatics and Computational genomics
– Social networks
– Biometrics
– Energy and smart grid

* PLENARY SPEAKERS
– Martin Wainwright, Univ. California at Berkeley, USA
– Francis Bach, INRIA, France (PASCAL2 Invited Speaker)
– Ali H. Sayed, Univ. California at Los Angeles, USA

* IMPORTANT DATES
– Full paper submission: May 7, 2012
– Notification of acceptance: June 25, 2012
– Advance registration before: July 31, 2012
– Camera-ready paper: August 15, 2012
– Conference: September 23-26, 2012

* SUBMISSION
Authors are invited to submit a double-column paper of up to 6 pages using the electronic submission procedure described at http://mlsp2012.conwiz.dk.
– Accepted papers will be published by IEEE Press and in electronic format.
– Selected papers from MLSP 2012 will be considered for a special issue of The International Journal of Neural Systems (JCR IF 4.2).

* FURTHER INFORMATION:
PDF file of the Call-for-Papers can be found at http://mlsp2012.conwiz.dk
For up-to-date information on the technical and social program for practical arrangements, please visit the conference website at http://mlsp2012.conwiz.dk

* DATA COMPETITION, supported by Amazon.com

A data and signal analysis competition is being organized. Winners will present their work and receive their award during the Workshop. The goal is to select/design a classifier (and any pre-processing systems, including a feature extractor) that correctly classifies a pair of an employee and a resource (a resource can be a computer, a data base, or a data portal etc) into one of two classes: access or non-access. The winner will be the submission that minimize the number of manual access grant/removal operations in the future testing data.
Eligibility: Anyone
Registration: Registration is not required. However, if you wish to receive important updates on the competition by email then please send a request to the address provided on the web page.
Deadline: Submissions must be emailed to the email address provided at the web page no later than one week after the main conference paper submission deadline, i.e. by May 14, 2012.
Awards: Up to three newest Kindle Fire tablet computers from Amazon will be awarded as prizes. Amazon will support and invite the winners to present their results at Amazon’s new SLU headquarter in Seattle, USA.
Data and more details: Available on the competition page: https://sites.google.com/site/amazonaccessdatacompetition/

* ORGANIZATION

General Chairs:
Ignacio Santamaría (Univ. Cantabria, Spain)
Jerónimo Arenas-García (Univ. Carlos III de Madrid, Spain)
Gustavo Camps-Valls (Univ. de València, Spain)

Technical Chairs:
Deniz Erdogmus
Fernando Pérez-Cruz (Univ. Carlos III de Madrid, Spain)

Special Sessions Chairs:
Emilio Parrado-Hernández (Univ. Carlos III de Madrid, Spain)
Jocelyn Chanussot (Grenoble Institute of Technology, France)

Data Competition Chairs:
Kenneth E. Hild II (Oregon Health & Science University, USA)
Vince Calhoun (University of New Mexico, USA)
Weifeng Liu (Amazon.com Inc., USA)

Publicity Chairs:
Marc Van Hulle (K.U. Leuven, Belgium)
Luis Gómez Chova (Univ. de València, Spain)

Web and Publication Chair:
Jan Larsen (Technical University of Denmark, Denmark)

Local Organizing Chairs:
Jesús Ibáñez (Univ. Santander)
Javier Vía (Univ. Santander)

ECML PKDD 2012 Discovery Challenge: Third Challenge on Large Scale Hierarchical Text Classification

Web site: http://lshtc.iit.demokritos.gr/
Email: lshtc_info@iit.demokritos.gr

This year’s discovery challenge hosts the third edition of the successful PASCAL challenges on large scale hierarchical text classification. The challenge comprises three tracks and it is based on two large datasets created from the ODP web directory (DMOZ) and Wikipedia. The datasets are multi-class, multi-label and hierarchical. The number of categories ranges between 13,000 and 325,000 roughly and the number of documents between 380,000 and 2,400,000.

The tracks of the challenge are organized as follows:

1. Standard large-scale hierarchical classification
a) On collection of medium size from Wikipedia
b) On a large collection from Wikipedia
2. Multi-task learning, based on both DMOZ and Wikipedia category systems
3. Refinement-learning
a) Semi-Supervised approach
b) Unsupervised approach

In order to register for the challenge and gain access to the datasets you
must have an account at the challenge Web site.

Important dates:

– March 30, start of the challenge
– April 20, opening of the evaluation
– June 29, closing of evaluation
– July 20, paper submission deadline
– August 3, paper notifications

Organizers
– Ion Androutsopoulos, AUEB, Athens, Greece
– Thierry Artieres, LIP6, Paris, France
– Patrick Gallinari, LIP6, Paris, France
– Eric Gaussier, LIG, Grenoble, France
– Aris Kosmopoulos, NCSR “Demokritos” & AUEB, Athens, Greece
– George Paliouras, NCSR “Demokritos”, Athens, Greece
– Ioannis Partalas, LIG, Grenoble, France

CALL FOR PARTICIPATION: ACM 3rd INTERNATIONAL SYMPOSIUM ON FACIAL ANALYSIS AND ANIMATION (FAA)

http://faa2012.ftw.at/

Museum of Young Art, Vienna. September 21st, 2012

Important Dates:
13th July 2012: Deadline for extended abstract submission
17th August 2012: Notification of acceptance
7th September 2012: Camera ready abstracts

Facial animation is a broad and exciting area of research drawing on multiple disciplines: computer graphics and animation provide the means to render and display a face, computer vision can be used to measure, interpret and decode facial actions, while psychology can help provide the emotive human element of animation. However, creating convincing facial animation is an exceptionally difficult task – each one of us is an expert judge in deciding whether an animation is realistic or not. In today’s world, facial animation has more applications than ever before: from video game characters to movie actor doubles, from machine facial displays to psychological research stimuli.

Following on from the success of FAA 2009 (http://www.cstr.ed.ac.uk/faa2009/), and FAA 2010 (http://www.cstr.ed.ac.uk/faa/), we are pleased to announce a call for submissions for FAA 2012, in cooperation with ACM. The aim of this meeting is to bring together researchers and practitioners from both academia and industry – particularly in VFX and games – interested in all aspects of facial animation and related analysis. Submissions are invited in the following broad topic areas:

• Acquisition of Facial Shape, Motion and Texture
• Performance Driven Animation and Expression Mapping
• Facial animation using Example Based Synthesis and Motion Graph based techniques
• Facial Animation Production Pipelines
• Visual Speech Synthesis
• Animation of Non-Linguistic Behaviors and Vocalisations
• Perception of Facial Animation and the “Uncanny Valley”
• Facial Rendering (Photorealistic and Non-Photorealistic)
• Virtual Characters for Telepresence / Embodied Virtual Agents
• Facial Model Based Coding and Compression
• Facial Analysis and Animation for Mobile Applications

Research should be submitted as a 1 page extended abstract using the SIGGRAPH formatting guidelines (see http://www.siggraph.org/publications/instructions). LaTeX and BibTeX class files following the “acmsiggraph” convention may be downloaded from (http://www.siggraph.org/publications/acmsiggraph.zip). An example submission may be found at (http://www.siggraph.org/publications/poster-abstract-example1.pdf).

Submissions do not need to be anonymous. Contributions should submitted following instructions on the FAA 2012 website. Authors of accepted submissions will have their work appear in the ACM Digital Library (http://portal.acm.org/dl.cfm). The program on the day will consist of both oral and poster presentations, including invited keynote talks from leading international experts.

Chairs: Darren Cosker (Uni. of Bath), Michael Pucher (FTW), Gregor Hofer (FTW), Michael Berger (Uni. of Edinburgh) and Will Smith (Uni. of York).

For general inquiries regarding FAA, including contact information for the organisers, please visit: http://faa2012.ftw.at/contact.html

PhD and Post Doc Position in NLU and Dialog

=======================================================
Two positions in the area
natural language understanding and human machine dialog
at Saarland University
=======================================================

We anticipate the availability for two positions in the area
of natural language understanding and dialog modeling,
one position for a PhD student and a second position
for a postdoctoral researcher or equivalently qualified
senior researcher.

The task is to conduct research in cross-domain natural
language understanding as well as generic dialog
modeling combining rule-based and statistical
techniques. The successful candidate should have a
degree in computer science, computational linguistics
or a discipline with a related background. Excellent
programming skills are important. A good math background
is a plus. Very good oral and written communication skills
in English are also required. The research will be carried
out together with a European consortium of high-profile
research institutes and companies.

Saarland University is one of the leading European research
sites in computational linguistics and offers an active,
stimulating research environment. Close working relationships
are maintained between the Departments of Computational
Linguistics and Computer Science. Both are part of the Cluster
of Excellence, which also includes the Max Planck Institutes
for Informatics (MPI-INF) and Software Systems (MPI-SWS) and
the German Research Center for Artificial Intelligence (DFKI).

Both positions are fully funded positions with a salary
in the range of 37,000 Euros to 51,000 Euros per year
depending on the qualification and professional experience
of the successful candidates. Starting date is
November 1st. The positions are for three years.

Each application should include:

* Curriculum Vitae including a list of publications
(if applicable)
* Transcript of records
* Short statement of interest (not more than half a
page)
* Names of two references
* Any other supporting information or documents

Applications (documents in PDF format in a single file)
should be sent no later than , Monday May 7th to:
Diana.Schreyer@LSV.Uni-Saarland.De

Further inquiries regarding the project should be
directed to:
Dietrich.Klakow@LSV.Uni-Saarland.De

Open PhD position in Machine Learning and Computer Vision (Idiap, Switzerland)

======================================================================
Open position in Machine learning and computer vision
======================================================================

The Idiap Research Institute, affiliated with École Polytechnique
Fédérale de Lausanne, seeks one PhD student in statistical learning to
develop original active-learning techniques to collect and leverage
very large training sets for image classification and object
detection.

http://www.idiap.ch/~fleuret/hiring.html

This position is funded by a grant from the Swiss National Science
Foundation, and the candidate will be a doctoral student at EPFL.

Research will be done under the supervision of Dr. François Fleuret.

* Summary

Recent advances in Machine Learning aim at leveraging very large
training sets. The objective of this project is to adapt
state-of-the-art supervised learning methods to use partially
annotated training sets, and to exploit structures known a priori on
the said sets to concentrate computation during learning on the most
difficult and informative subsets of data.

The application domain will be image classification and object
detection in natural images. This work will mix theoretical
developments in statistical learning with the implementation of
algorithms working on real-world data.

Applicants must be self-sufficient programmers and have a strong
background in mathematics. They should be familiar with several of the
following topics: probabilities, applied statistics, information
theory, signal processing, optimization, algorithmic, and C++
programming.

* About Idiap

The Idiap Research Institute is located in Valais, a scenic region in
the south of Switzerland, surrounded by the highest mountains of
Europe, and within close proximity to Lausanne and Geneva. The working
language of Idiap is English.

Please contact francois.fleuret@idiap.ch for additional information.

CFP ICML workshop 2012: Sparsity, Dictionaries and Projections in Machine Learning and Signal Processing

­====================================================================
CALL FOR PAPERS

Sparsity, Dictionaries and Projections in Machine Learning and Signal
Processing
ICML 2012 Workshop, Edinburgh, Scotland
30 June or 1 July, 2012

http://www.di.ens.fr/~obozinski/ICML2012workshop/

Submission Deadline: Monday, May 7, 2012 (5:00 pm PDT)
====================================================================
Overview

Sparse representations are today key in many fields of applied
mathematics faced with data, from signal processing to machine
learning and statistics.

Historically, several communities proposed various approaches to
sparse coding: on the one hand the use of carefully crafted
dictionaries, like wavelets, forming bases of functional spaces with
good approximation properties over a class of signals, but constructed
without data; on the other hand, the use of representations derived
directly from data via either algebraic formulations like sparse
matrix factorization, or via probabilistic formulations, based on the
introduction of latent variables, such as, e.g., independent component
analysis or latent Dirichlet allocation.

The introduction of sparsity and/or structure in matrix factorization
scheme, which where previously used for dimensionality reduction,
induced major shifts in several existing paradigms and led to
significant breakthroughs, which have demonstrated the ability of
sparse models to provide concise descriptions of certain high
dimensional data through low-dimensional projections, together with
algorithms of provable performance and bounded complexity. Compressed
sensing (and more generally the clever use of random low-dimensional
projections), dictionary learning, and non-parametric topic models,
are just a few of the rapidly emerging paradigms in this area at the
confluence of signal processing and machine learning.

While sparse models and random low-dimensional projections are already
at the heart of several success stories in signal processing and
machine learning, their full potential is yet to be achieved and calls
for further understanding. The goal of the workshop is to confront the
various point of views and foster exchanges of ideas between the
signal processing, statistics, machine learning and applied
mathematics communities.

We encourage submissions exploring various aspects of learning sparse
models and/or latent representations, in the form of new algorithms,
theoretical advances and/or empirical results. Some specific areas of
interest include structured matrix factorization algorithms, Bayesian
models for latent variable representations, analysis of random
dictionaries versus learned dictionaries, novel applications of
dictionary learning or relationships to compressed sensing.

Submission Guidelines

Submissions should be written as extended abstracts, no longer than 4
pages in the ICML latex style. Style files and formatting instructions
can be found at http://icml.cc/2012//files/icml2012stylefiles.zip.
Submissions must be in PDF format. Authors’ names and affiliations
should be included, as the review process will not be double blind.
The extended abstract may be accompanied by an unlimited appendix and
other supplementary material, with the understanding that anything
beyond 4 pages may be ignored by reviewers.

Please send your PDF submission by email to guillaume.obozinski@ens.fr
with the words “ICML workshop submission” in the title by 5:00 pm PDT
on Monday, May 7. Notifications will be given on or before 21 May.
Please include these words “ICML workshop” in the title of mails you
are sending about the workshop or workshop submissions. Work that is
pending review, was recently published or was presented elsewhere will
be considered, provided that the extended abstract mentions this
explicitly. Finally, note that there will be no official proceedings
from this workshop.

Invited Speakers (to be confirmed)

Pierre Comon (CNRS / University Nice Sophia Antipolis)
Julien Mairal (University of California at Berkeley)
Matthias Seeger (Ecole Polytechnique Federale de Lausanne)
Daniel Vainsencher (Technion – Israel Institute of Technology)

Organizers

Michael Davies (Edinburgh), Rémi Gribonval (INRIA), Rodolphe Jenatton
(CNRS) and Guillaume Obozinski (INRIA)