Two Chairs of Statistics at UCL

The Department of Statistical Science at UCL is currently undergoing a
programme of expansion. As part of this expansion the Department invites
applications for two Chairs of Statistics from candidates with outstanding
research records of international excellence. The two appointees will be
expected to provide research leadership as well as to contribute to the
Department’s undergraduate and postgraduate teaching programmes. An
initial task will be to provide input into the recruitment of two new
lecturers. More experienced candidates may be considered for the future
Headship of the Department.

The successful candidates will have internationally recognised track
records of research and will be able to contribute to the strong research
profile of the Department. They are expected to have continuing track
records of external research grant funding and doctoral research
supervision, experience of teaching at undergraduate or postgraduate level
and excellent interpersonal, oral and written communication skills.

Both posts are available from 1 October 2011. Salary is negotiable on the
professorial scale but will not be less than £61,960 per annum inclusive
of London Allowance. Provision of an additional market supplement on top
of the basic salary is currently under review.

Further particulars including a job description, person specification and
details of how to apply can be accessed at
http://www.ucl.ac.uk/statistics/department/jobs. Informal enquiries may be
addressed to Professor Valerie Isham, Head of Department, tel.
+44(0)20-7679-1868, email valerie(at)stats.ucl.ac.uk.

We particularly welcome female applicants and those from an ethnic
minority, as they are currently under-represented within UCL at this
level.

Closing date: 26 June 2011

Interviews date: interviews to be held in July 2011

SOR 11 – PASCAL2 sponsored conference announcement and call for papers

The 11th International Symposium on Operations Research in Slovenia (SOR’11) will be held in Dolenjske Toplice, Slovenia, September 28-30, 2011. The professional objectives of the international symposium are to provide a forum for international and national exchange of ideas on various aspects of OR and related fields. http://sor11.fis.unm.si/home

Topics
– Professional aspects of OR (OR methodology, OR education, OR implementation, OR profession).
– Methods and techniques of OR (Assignment, Combinatorial optimization, Decision theory, Games, Integer programming, Linear programming, Multiple criteria decision making, Networks and graphs, Non-linear programming, Numerical methods, Simulation, Statistics, Stochastic processes, Vector optimization, etc.
– Areas of application (Agriculture, Banking, Ecology, Economic systems and econometrics, Energy, Environmental protection, Finance, Inventory, Production planning, Transport, etc.).
– Information and computing aspects of OR (Artificial intelligence, Decision support systems, Expert systems, Information systems, OR software, etc.).

In addition, there is a PASCAL2 special session open in the submission page. Members conference fee applies to all PASCAL2 researchers.
Deadline for Submission of contributed papers June 15, 2011 (Extended dealine)

Should you need assistance, or you have any questions about the SOR’11, please feel free to contact: sor11(at)fis.unm.si.

Open PhD Position in ML at Grenoble (France)

* Large Scale Text Categorization
* Grenoble University, Grenoble, France
* Position open from 1 September 2011 (earlier or later start dates are negotiable)

PROJECT DESCRIPTION

Several categorization problems rely on category systems comprising several thousands of categories, in which new elements have to be placed. For example, DMOZ (http://dmoz.org/), which is supposed to be the largest repository of the web, contains more than 590,000 categories, in which new web pages are categorized by a team of worldwide volunteers, each being in charge of a sub-portion of the system. A similar situation is faced with systems aiming at assigning concepts or tags, from a given ontology/thesaurus or sets of tags, to documents or parts of documents. PubMed, for example, contains more than 16 million references, the abstracts of which are indexed with concepts from MeSH (http://www.nlm.nih.gov/pubs/factsheets/mesh.html), a thesaurus comprising more than 150,000 concepts. Similarly, social networks à la Flickr make use of tags which are manually assigned by users, but which can be automatically “propagated” to ensure consistence over the entire network. In all these examples, the number of categories involved requires, for both maintaining and browsing the category system, a hierarchical organization of the categories.

Several approaches have been investigated for deploying categorizers in large-scale category systems, as big-bang approaches, in which one tries to directly categorize documents into the leaf categories, without making use of the category hierarchy, or top-down approaches, in which the hierarchy is exploited to divide the overall classification problem into smaller ones. Studies reported in [1] and [2] suggested that the direct deployment of a single classifier onto the complete set of categories (big-bang approaches) may lead to training and classification time which are unacceptable. On the contrary, top-down approaches, by simplifying the overall problem, lead to acceptable classification times. However they tend to propagate errors made at higher levels of the hierarchy. More recently, alternative approaches have been proposed, aiming at locally propagating tags to close neighbors according to the tags already present and the topology of the network [3], or attempting to find efficient representations of large scale collections (either of documents, as the semantic hashing approach proposed in [4], or of the category system, as the conditional probability trees defined in [5]).

We want to study these different approaches, and in particular devise methods to obtain the optimal subset of categories to be considered at any given point in a tree or cascade of categorizers. This is related to the two following questions:

1. What happens on the generalization error of the classifier, when we replace a category by its children?
2. What is the optimal (according to the generalization error) subset of categories to retain, and how to efficiently represent them and deploy categorizers?

We want to investigate these points with several types of classifiers (as discriminative, large margin classifiers, or maximum likelihood classifiers), and develop theoretically well founded methods which are proved effective on very large data collections, as the ones currently used in the Large Scale Hierarchial Text Categorization Challenge of the PASCAL network of excellence (http://lshtc.iit.demokritos.gr:10000/LSHTC2_CFP).

*Keywords: classification, large set of categories, conditional probability trees

*Bibliography*

[1] T.-Y. Liu, Y. Yang, H. Wan, H.-J. Zeng, Z. Chen, W.-Y. Ma. Support vector machines classification with a very large-scale taxonomy, SIGKDD Explorations, 2005.
[2] O. Madani, W. Greiner, D. Kempe, M. R. Salavatipou. Recall systems: Efficient learning and use of category indices, International Conference on Artificial Intelligence and Statistics (AISTATS), 2007.
[3] S. Bao, B. Yang, B. Fei, S. Xu, Z. Su, Y. Yu. Boosting social annotations using propagation, Proceeding of the 17th ACM conference on Information and knowledge management (CIKM ’08), 2008.
[4] R. Salakhutdinov, G. Hinton. Semantic hashing, International Journal of Approximate Reasoning, 2009.
[5] A. Beygelzimer, J. Langford, Y. Lifshits, G. Sorkin, A. Strehl. Conditional Probability Tree Estimation Analysis and Algorithms, 2009 (arXiv:0903.4217v2).

REQUIREMENTS

The candidate for the PhD position should have a Master’s degree in computer science or applied mathematics with knowledge of machine learning concepts. The ideal candidate should be able to conduct theoretical research, but also implement and test models on very large datasets.

APPLICATION

Applicants should send (preferably as a single PDF file):

* a CV
* a brief statement of research interests
* references (with email and phone number)
* their academic transcript
* a sample of strongest publications or course work (e.g. Master thesis)

Applications and inquiries should be directed to:

Cecile Amblard – cecile.amblard(at)imag.fr
Eric Gaussier – eric.gaussier(at)imag.fr

*Duration*

3 years (starting in Sept. 2011)

The PhD is fully financed through a French Research National Agency grant.

BBC-Surrey PhD Industrial Case Studentship: Computer Vision

PhD in Computer Vision and Graphics
BBC-University of Surrey
Industrial PhD Case Studentship

42 months PhD funding of fees(UK/EU) + stipend circa. £20K/annum (tax-free)

(£14K/annum EPSRC stipend+ £6K/annum BBC sponsorship)

This EPSRC Industrial PhD case studentship will support the successful applicant to conduct research at both BBC R&D and the University of Surrey taking advantage of existing expertise and state-of-the-art facilities. The proposed PhD research will develop expertise in advanced computer vision, video analysis and computer graphics techniques, together with gaining experience of their application in broadcast and game industries.

Further details: http://cvssp-data.eps.surrey.ac.uk/Personal/AdrianHilton/Vacancies.html

ALL applicants must be UK residents.

Interested applicants should contact Dr. Oliver Grau (oliver.grau(at)bbc.co.uk) or Prof. Adrian Hilton (a.hilton(at)surrey.ac.uk) for further information. To apply for the industrial PhD studentship please send a full CV to Prof. Adrian Hilton.

Closing date for applications: Wednesday 1st June 2011

Starting date: must start on or before October 1st 2011

ICML Workshop on Unsupervised and Transfer Learning

July 2, 2011
Bellevue, Washington state, USA

http://clopinet.com/isabelle/Projects/ICML2011/

Call for breaking news posters and demos
Send abstract before June 20, 2011 at utl(at)clopinet.com

You are cordially invited to the lunch, which will be served during the poster session.

Invited speakers:
Pierre Baldi, UC Irvine, California, USA
Yoshua Bengio, Universite de Montreal, Canada
Joachim Buhmann, ETH, Zurich, Switzerland
Alexandru Niculescu-Mizil, NEC labs, New Jersey, USA
Gunnar Raetsch, MPI, Germany
Dale Schuurmans, University of Alberta, Canada
Prasad Tadepalli, Oregon State University, Oregon, USA
Ruslan Salakhutdinov, MIT, Massachussetts, USA
Qiang Yang, Hong Kong University of Science and Technology

Organizers:
Isabelle Guyon, Clopinet, Berkeley, California
Daniel Silver, Acadia University, Canada

Vision and Sports Summer School 2011 – registration closes 25 May

Vision and Sports Summer School 2011
Zurich, 22-26 August 2011
http://www.vision.ee.ethz.ch/summerschool2011/
email: vs3(at)vision.ee.ethz.ch
application deadline: 25 May 2011

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 Indoor
Climbing, Judo, Tennis and Volleyball. 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).

TEACHERS

Jiri Matas
Czech Technical University

Daniel Cremers
TU Muenchen

Vittorio Ferrari
ETH Zurich

Silvio Savarese
University of Michigan

Christoph Lampert
IST Austria

Patrick Perez
Technicolor Research and Innovation

Cristian Sminchisescu
University of Bonn

Bodo Rosenhahn
University of Hannover

Ondrej Chum
Czech Technical University (Prague)

Lubor Ladicky
University of Oxford

COMPUTER VISION LECTURES

Current list of topics:

Local feature extraction
Multi-view geometry and 3D reconstruction
Large-scale specific object recognition
Appearance-based object categorization
Shape representation and matching
Contour-based object categorization
Human Pose Estimation
Kernel Methods for Computer Vision
Continuous optimization for Computer Vision
Structured Probabilistic models in Computer Vision
Markov Random Fields and Conditional Random Fields for Computer Vision
Tracking in video

SPORT ACTIVITIES

Tennis, Volleyball, Indoor Climbing, Unihockey, Table Tennis, Soccer,
Judo, Basketball

APPLICATION

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

http://www.vision.ee.ethz.ch/summerschool2011/

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
Master/PhD students is 400 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.

Master/PhD applicants from the United States will automatically be
considered for a grant to cover part of their travel expenses.

Applicants should apply before 25 May 2011.
Notification of acceptance will be sent by 31 May 2011.

MORE INFORMATION

http://www.vision.ee.ethz.ch/summerschool2011/

NIPS 2011 Call For Demonstrations

Demonstration Proposal Deadline: Monday September 19, 2011
The Neural Information Processing Systems Conference 2011
http://nips.cc/Conferences/2011/ has a Demonstration Track running in
parallel with the evening Poster Sessions, December 13-15, 2011, in
Granada, Spain.

Demonstrations are an opportunity to showcase:
* Hardware technology
* Software systems
* Neuromorphic and biologically-inspired systems
* Robotics
or other systems, which are relevant to the technical areas covered by NIPS
(see Call for Papers http://nips.cc/Conferences/2011/CallForPapers) .
Demonstrations must show novel technology and must be live, preferably with
some interactive parts. A demonstration is not just another poster
presentation or a slide show, the action part is important.

Submissions: Submission of demo proposals at the following URL:
https://nips.cc/Demonstrators/

You will be asked to fill a questionnaire and describe clearly:
* the technology demonstrated
* the elements of novelty
* the live part
* the interactive part
* the equipment brought by the demonstrator
* the equipment required at the place of the demo

Evaluation Criteria: Submissions will be refereed on the basis of technical
quality, novelty, live action, potential for interaction.

Demonstration chair: Samy Bengio
Call URL: http://nips.cc/Conferences/2011/CallForDemonstrations

NIPS 2011 Call For Workshops

Natural and Synthetic NIPS*2011 Post-Conference Workshops
December 16 and 17, 2011 Hotel Meliá Sol y Nieve and Sierra Nevada,
Granada, SPAIN

Following the regular program of the Neural Information Processing Systems
2011 conference in Granada, Spain, workshops on a variety of current topics
in neural information processing will be held on December 16 and 17, 2011,
in Sierra Nevada, Spain. We invite researchers interested in chairing one
of these workshops to submit proposals for workshops. The goal of the
workshops is to provide an informal forum for researchers to discuss
important research questions and challenges. Controversial issues, open
problems, and comparisons of competing approaches are not only encouraged
but preferred as workshop topics. Representation of alternative viewpoints
and panel-style discussions are also particularly encouraged.

Potential workshop topics include, but are not limited to:
Active Learning, Attention, Audition, Bayesian Networks, Bayesian
Statistics, Benchmarking, Biophysics, Brain-Machine Interfaces, Brain
Imaging, Cognitive Neuroscience, Computational Biology and Bioinformatics,
Computational Complexity, Control, Genetic/Evolutionary Algorithms, Graph
Theory, Graphical Models, Hippocampus and Memory, Human-Computer
Interfaces, Implementations, Kernel Methods, Mean-Field Methods, Music,
Natural Language Processing, Network Dynamics, Neural Coding, Neural
Plasticity, Neuromorphic Systems, On-Line Learning, Optimization,
Perceptual Learning, Robotics, Rule Extraction, Self-Organization, Signal
Processing, Social Networks, Spike Timing, Speech, Supervised/Unsupervised
Learning, Time Series, Topological Maps, and Vision.

Detailed descriptions of previous workshops may be found at:
http://nips.cc/Conferences/2010/Program/schedule.php?Session=Workshops

There will be seven hours of workshop meetings per day, split into morning
and afternoon sessions, with free time between the sessions for ongoing
individual exchange or outdoor activities. Selected workshops may be
invited to submit proceedings for publication in the post-NIPS workshops
monographs series published by the MIT Press.

Workshop organizers have several responsibilities, including:
*Coordinating workshop participation and content, including arranging short
informal presentations by experts, arranging for expert commentators to
sit on discussion panels, formulating discussion topics, etc.
*Providing the program for the workshop in a timely manner for the
workshop booklet.
*Moderating the discussion, and reporting its findings and conclusions to
the group during the evening plenary sessions.
*Writing a brief summary and/or coordinating submitted material for
post-conference electronic dissemination.

Submission Instructions
A nips.cc account is required to submit the Workshops application. Please
follow the url below and check the required format for the application well
before the deadline for workshop proposals. You can edit your application
online right up until the deadline.

Interested parties must submit a proposal by 23:59 UTC on July 1st, 2011.
Proposals should be submitted electronically at the following URL:

https://nips.cc/Workshops/

Preference will be given to workshops that reserve a significant portion of
time for open discussion or panel discussion, as opposed to a pure
“mini-conference” format.

We suggest that organizers allocate at least 50% of the workshop schedule
to questions, discussion, and breaks. Past experience suggests that
workshops otherwise degrade into mini-conferences as talks begin to run
over. For the same reason, we strongly recommend that each workshop include
no more than 12 talks per day.

This year we’d like to attempt to partially unify the NIPS workshop
important dates across all of the workshops. Therefore, please consider
using the following date guidelines for your workshop:
* Your workshop call should be publicized on or before August 30th, 2011.
* Submission deadline should be on or before September 23rd, 2011.
* Acceptance decisions mailed out on or before October 15th, 2011.

We stress that these are not mandatory, rather suggestions. If there are
circumstances that would make your workshop difficult using these dates,
you may use other dates. Also a call for contributions is not required and
is orthogonal to the decision about workshop acceptance.

NIPS does not provide travel funding for workshop speakers. In the past,
some workshops have sought and received funding from external sources to
bring in outside speakers. In any case, the organizers of each accepted
workshop can name two individuals to receive free registration for the
workshop program.

Jeff Bilmes and Fernando Perez-Cruz
NIPS*2011 Workshops Chairs
Call URL: http://nips.cc/Conferences/2011/CallForWorkshops

Call for breaking news posters and demos – Gesture Recognition

CVPR Workshop on Gesture Recognition
and launching of a benchmark program
June 20, 2011
Colorado Springs, USA
http://clopinet.com/isabelle/Projects/CVPR2011/

Call for breaking news posters and demos
Send abstract before June 10, 2011

Invited speakers:
Aleix Martinez, Ohio State University, USA.
Greg Mori, Simon Fraser University, Canada.
Richard Bowden, Univ. Surrey, UK.
Graham Taylor, NYU, New-York.
David Forsyth, University of Illinois at Urbana-Champaign.
Sudeep Sarkar, University of South Florida.
Dimitri Metaxas, Rutgers University, New Jersey.
Maja Pantic, Imperial College, London.
Christian Vogler, Institute for Language and Speech Processing, Athens, Greece.
Takeo Kanade, Carnegie Mellon University.

Organizing committee:
Isabelle Guyon, Clopinet, Berkeley, California
Vassilis Athitsos, University of Texas at Arlington
Jitendra Malik, UC Berkeley, California
Ivan Laptev, INRIA, France

Call for participation: ICANN 2011 – Machine learning re-inspired by brain and cognition

Call for Participation: ICANN 2011

The Twentieth Anniversary ICANN is back at its roots:
Machine learning re-inspired by brain and cognition

International Conference on Artificial Neural Networks
14 – 17 June 2011, Espoo, Finland
http://www.cis.hut.fi/icann2011

Registration is now open.

—

The International Conference on Artificial Neural Networks (ICANN)
is the annual flagship conference of the European Neural Network
Society (ENNS). In 2011, ICANN returns to its roots after 20 years.
The very first ICANN in 1991 was organized at Helsinki University
of Technology in Espoo, Finland. We invite all neural network
researchers worldwide to join us in celebrating this 20th
anniversary of ICANN and to see the latest advancements in our fast
progressing field.

ICANN 2011 presents research on two major themes, Brain-inspired
computing and Machine learning research; Keynote speakers and
competitions highlight cross-disciplinary interactions and
applications.

VENUE

ICANN 2011 will be held in the Dipoli Congress Center located on
the beautiful campus of Aalto University (former Helsinki
University of Technology), in Espoo (8km west from the city centre
of Helsinki). The time of the year is particularly suitable for
visiting Finland.

CONFERENCE TOPICS

ICANN 2011 presents research on two major themes, Brain-inspired
computing and Machine learning research, with strong cross-
disciplinary interactions and applications. A non-exhaustive list of
topics:

– Brain inspired computing: Connectionist cognitive science, Neural
and hybrid architectures and learning algorithms, Neural control
and planning, Reinforcement learning, Computational neuroscience,
Neural dynamics and complex systems, Self-organization, Neuro-
cognitive architectures, Recurrent networks

– Machine learning research: Graphical models, Bayesian networks,
Kernel methods, Generative models, Information theoretic
learning, Nonlinear projection, Relational learning, Online
learning, Dynamical models, Reinforcement learning

– Applications and cross-disciplinary connections: Data analysis,
Pattern recognition, Signal and time series processing, Blind
source separation, Hardware implementations and embedded systems,
Intelligent multimedia, Knowledge management, Multimodal
interfaces, Vision and image processing, Biomedical image
analysis, Speech and language processing, Robotics applications,
Intelligent control, Neuroinformatics, Bioinformatics, Biomedical
applications, Brain-computer interfaces, Critical infrastructure
systems, Complex networks

PLENARY SPEAKERS

Tom Griffiths, University of California Berkeley
(http://cocosci.berkeley.edu/tom/)

Riitta Hari, Aalto University
(http://ltl.tkk.fi/wiki/Riitta_Hari)

Geoffrey Hinton, University of Toronto
(http://www.cs.toronto.edu/~hinton/)

Aapo Hyvarinen, University of Helsinki
(http://www.cs.helsinki.fi/u/ahyvarin/)

John Shawe-Taylor, University College London
(http://www.cs.ucl.ac.uk/staff/j.shawe-taylor/)

Josh Tenenbaum, Massachusetts Institute of Technology
(http://web.mit.edu/cocosci/josh.html)

All ICANN participants are also welcome to follow the WSOM
2011 plenary session by Teuvo Kohonen on “Contextually
Self-Organized Maps of Chinese Words”.

ORGANIZATION

General chair: Erkki Oja

Program chairs: Wlodzislaw Duch, Mark Girolami,
Timo Honkela, Samuel Kaski

Workshop chair: Alexander Ilin

Local chair: Amaury Lendasse

Publicity chair: Jaakko Peltonen

Organizing committee members: Francesco Corona, Krista Lagus,
Yoan Miche, Ilari Nieminen,
Mari-Sanna Paukkeri, Tapani Raiko,
Ricardo Vigario

AREA CHAIRS

Peter Auer, Austria
Christian Bauckhage, Germany
Wray Buntine, Australia
Vince Calhoun, USA
Antonius Coolen, UK
Barbara Hammer, Germany
Giulio Jacucci, Finland
Kristian Kersting, Germany
Mikko Kurimo, Finland
Neil Lawrence, UK
Te-Won Lee, USA
Hiroshi Mamitsuka, Japan
Fernando Morgado Dias, Portugal
Klaus-Robert Muller, Germany
Klaus Obermayer, Germany
Cheng Soon Ong, Switzerland
Jan Peters, Germany
Marios Polycarpou, Cyprus
Jose Principe, USA
Volker Roth, Switzerland
Craig Saunders, UK
Alan Stocker, USA
Masashi Sugiyama, Japan
Ron Sun, USA
Peter Tino, UK
Alfred Ultsch, Germany
Koen Van Leemput, USA
Michel Verleysen, Belgium
Jean-Philippe Vert, France
Ole Winther, Denmark
Chang D. Yoo, South Korea

PROGRAMME

ICANN 2011 features six plenary sessions, a panel discussion on
central issues related to neural networks and machine learning
research, five workshops, fourteen oral sessions, a poster spotlight
session, and a poster session.

The full programme of ICANN 2011 is now available at
http://www.cis.hut.fi/icann2011 .

WORKSHOPS

ICANN 2011 hosts five workshops:
– Computational Intelligence for Quality of Life Environmental
Information Services
– Validation of Computational Models in Social and Economic Sciences
– Beyond Correlations: Developments in Supervised Learning Algorithms
for Spiking Neural Networks
– Challenge Workshop: Mind Reading Competition on MEG Data
– META-NET Workshop: Context in Machine Translation

Details of the workshops will be available at
http://www.cis.hut.fi/icann2011 .

WSOM 2011, the 8th Workshop on Self-Organizing Maps (13-15 June
2011) will be co-located with ICANN 2011.

COMPETITIONS

Mind reading competition on MEG data: classify from MEG signals
which type of video stimulus the subject is viewing such as football
match, movie, natural scenery, etc.

META-NET Multimodal Machine Translation Challenge: choose the
best translation from translations given by multiple machine
translation systems, using additional context information like
domain, surrounding text, etc.

See http://www.cis.hut.fi/icann2011 for more details.

REGISTRATION

Registration for ICANN 2011 is now open. The on-line registration
is available at http://www.cis.hut.fi/icann2011 .

In addition to ICANN registration, joint registration to both ICANN
2011 and WSOM 2011 is available. See http://www.cis.hut.fi/icann2011
for the full range of registration options and prices.

ACCOMMODATION

A conference price has been arranged at the Sokos Presidentti Hotel,
and convenient public transportation to the conference site is
available. See http://www.cis.hut.fi/icann2011 for accommodation
details.

SPONSORS

ICANN 2011 is supported by European Neural Network Society (ENNS),
Pattern Recognition Society of Finland and Finnish Artificial
Intelligence Society.

====== See http://www.cis.hut.fi/icann2011 for more details. ======