Workshop: “Sparse structures: statistical theory and practice”, 16-18 June 2010, Bristol, UK

SPARSE STRUCTURES: STATISTICAL THEORY AND PRACTICE
Research workshop: 16-18 June 2010, Bristol, UK

One of the most challenging problems in modern statistics is to find
effective methods for the analysis of complex multi-dimensional data.
Sparsity has been emerging as a major tool for handling statistical
problems in high dimensions. The aim of this workshop is to bring
together theory and practice in modelling high-dimensional data using
these tools.

Invited speakers

Christophe Ambroise (CNRS, Paris)
Francois Caron (Bordeaux)
Sara van de Geer (Zurich)
David Madigan (Columbia)
Nicolai Meinshausen (Oxford)
Sylvia Richardson (Imperial)
Ya’acov Ritov (Jerusalem)
Bernard Silverman (Oxford)
Mike Tipping (consultant)
Alexandre Tsybakov (Paris VI)
Martin Wainwright (Berkeley)

The workshop will also feature a contributed and poster programme.

The number of participants we can accept is limited, and we ask you to
write to us in advance if you wish to attend. We will have some
limited funds for partial support of expenses for junior researchers
(graduate students, and researchers within 3 years of receiving their
Ph.D. degree).

Deadlines:
For contributed talks: 15 February 2010 (!)
For poster or participation: 15 March 2010

Further information about the workshop can be found at
http://www.sustain.bris.ac.uk/ws-sparsity/

Ph.D. position INRIA Lille

A PhD studentship is available in at Sequential Learning
Lab in INRIA Lille, France.
The topic is time series prediction, and non-parametric statistical
analysis of time series.

Position:
3 year fully funded position (contract).
The position will start in September 2010. The deadline for applications
will be 1st of May, but interested candidates are encouraged to contact me earlier.

Environment:
The PhD student will work with Daniil Ryabko http://daniil.ryabko.net ,
and with other members of the SequeL group
http://sequel.futurs.inria.fr/ at INRIA Lille.
SequeL is one of the most dynamic labs at INRIA, with over 20
researchers (including PhD students)
working on both fundamental and practical aspects of sequential learning
problems:
from statistical learning, through reinforcement learning, to computer
poker and Go players.
INRIA is France’s leading institution in Computer Science, with over
2800 scientists
employed, of which around 250 in Lille.

Lille is the capital of the north of France, a metropolis with 1 million
inhabitants,
with excellent train connection to Brussels (40 min), Paris (1h) and
London (1.5h by train).

Subject:
The basic problem setup is as follows. There is an unknown stochastic
source of data, generating observations in a sequential fashion. The
data can be anything from stock market observations, to DNA sequences,
to behavioural sequences. There are several learning and inference
problems connected with it, of which the two most basic ones are:
predicting the probabilities of the next observations, and testing
hypotheses about the source (such as independence, homogeneity, etc.)
To solve these problems, one has to consider models of the data.
Different types of data require different models.

The goal is to describe those probabilistic models under which
successful learning is possible, for the inference problems
considered: sequential prediction, and hypothesis testing. This would
eventually lead to an automated modelling algorithms for sequential
learning. The primary goal, however, is to establish a theoretical
understanding of what is possible to learn, in the sequential problems
of interest considered, under which assumptions. More precisely, a data
source is a probability distribution on the set of all possible
sequences of observations, and a model is a set of such probability
distributions. We are interested in identifying the properties of models
which ensure the existence of efficient algorithms that are successful
(e.g. as predictors) given the model.

Requirements:

The successful candidate will have a MSc or equivalent degree in
mathematics or computer science,
with strong background in probability and statistics.
Programming skills will be considered a plus.
The working language in the lab is English.

For further information please email daniil.ryabko(at)inria.fr ,
with subject -SeqPHD-, joining a CV and a description of interests.

More information (including the application procedure) will soon be
available through
http://www.inria.fr/travailler/opportunites/doc.en.html

INRIA Visual Recognition and Machine Learning Summer School

INRIA Visual Recognition and Machine Learning Summer School

Grenoble, 26-30 July 2010

http://www.di.ens.fr/willow/events/cvml2010

Application deadline: 15 May 2010

OVERVIEW: The objective of this summer school is to provide an
overview of some of the latest advances in visual recognition together
with the related machine learning algorithms. It will take place at
the INRIA campus in Grenoble. The courses will be given by experts in
the field and complemented by a half-day hiking trip to the
surrounding mountains.

Lecturers F. Bach (INRIA)
D. Forsyth (UICU) Z. Harcahoui (INRIA)
C. Lampert (IST Austria) I. Laptev (INRIA)
C. Rother (Microsoft) C. Schmid (INRIA)
J. Ponce (ENS) J. Sivic (INRIA)
A. Zisserman (Oxford Univ.) J. Verbeek (INRIA)

Workshop on Automated Knowledge Base Construction 2010

First workshop on Automated Knowledge Base Construction

May 17-19, 2010

Grenoble, France

http://akbc.xrce.xerox.com

CALL FOR PAPERS

4-8 page submissions due March 5, 2010 for non-Pascal members

4-8 page submissions due March 15, 2010 for Pascal members

Good decision-making is dependent on comprehensive, accurate knowledge. But the information relevant to many important decisions in areas such as business, government, medicine and scientific research is massive, and growing at an accelerating pace. Relevant raw data is widely available on the web and other data sources, but usually in order to be useful it must be gathered, organized, and normalized into a knowledge base.

Hand-built knowledge bases such as Wikipedia have made us all better decision-makers. However more than human editing will be necessary to create a wide variety of domain-specific, deeply comprehensive, more highly structured knowledge bases.

A variety of automated methods have begun to reach levels of accuracy and scalability that make them applicable to automatically constructing useful knowledge bases. These capabilities have been enabled by research in areas including information extraction, information integration, databases, search and machine learning.

There are substantial scientific and engineering challenges in advancing and integrating such relevant methodologies.

This workshop will gather researchers in a variety of fields that contribute to the automated construction of knowledge bases.

There has recently been is a tremendous amount of new work in this area, some of it in traditionally disconnected communities. In this workshop the organizers aim to bring these communities together.

Topics of interest include:

* information extraction; open information extraction, named entity extraction; entity resolution, relation extraction.

* information integration; schema alignment; ontology alignment; ontology construction.

* alignment between knowledge bases and text.

* joint inference between text interpretation and knowledge base pattern analysis, reading the web, learning by reading.

* databases; distributed information systems; probabilistic databases.

* scalable computation; distributed computation.

* information retrieval; search on mixtures of structured and unstructured data; querying under uncertainty.

* machine learning; unsupervised, lightly-supervised and distantly-supervised learning; learning from naturally-available data.

* human-computer collaboration in knowledge base construction; automated population of wikis.

* inference; scalable approximate inference.

* languages, toolkits and systems for automated knowledge base construction.

* demonstrations of existing automatically-built knowledge bases.

Speakers / Participants include:

Philip Bohannon, Yahoo! Research

Michael Cafarella, University of Michigan

AnHai Doan, University of Wisconsin

Patrick Gallinari, LIP6

Alon Halevy, Google

Zachary Ives, University of Pennsylvania

Ron Kaplan, Powerset / Microsoft

Andrew McCallum, University of Massachusetts

Zaiqing Nie, Microsoft Research

Fernando Pereira, Google & University of Pennsylvania

Sunita Sarawagi, Indian Institute of Technology

Gerhard Weikum, Max Planck Institute

Important dates:

* Paper submission deadline: Friday 5 March 2010

* Notification of acceptance: Friday 2 April 2010

* Camera-ready due from authors: Monday 3 May 2010

* Workshop: Mon-Wed 17-19 May 2010 (just after AISTATS)

Organizing committee:

Program Chair: Andrew McCallum

Publication Chair: Guillaume Bouchard

Funding Chair: Cedric Archambeau

Local Arrangements: Onno Zoeter

Host Arrangements: Jean-Marc Andreoli

Workshop venue:

The workshop will take place in the chateau at Xerox Research Centre Europe, near Grenoble, France. In addition to technical talks, talks about submitted papers and a poster session, we will have plenty of time for informal discussions and community building: a reception dinner the evening before we begin, lunches provided on-site with time to talk, a banquet dinner on the first evening, and on the second day a half-day hike in the local Alps, including a raclette dinner at a refuge.

Postdoctoral position available: machine learning/cancer bioinformatics

Postdoctoral Position in Machine Learning and Cancer Bioinformatics
University of Bristol, United Kingdom (vacancy ref. 15239)

We are seeking to appoint an outstanding postdoctoral researcher interested
in machine learning and bioinformatics to work with Dr. Colin Campbell,
University of Bristol, United Kingdom. This project will involve the
design of new algorithms and data analysis techniques in addition to
an important core application study. We will use a variety of methods
from modern machine learning including Bayesian techniques and probabilistic
graphical methods, kernel-based methods and other approaches. The core
application study will involve the use of unsupervised, semi-supervised,
biclustering and other machine learning and statistical methods to delineate
subtypes of cancer and identify dysregulated genes within these subtypes.
We are also very interested in the development and use of data integration
methods which can handle multiple types of data (e.g. expression array,
microRNA array, sequence data) within the same model.

Aside from this core application program, the appointee will have
the freedom to pursue research in machine learning, bioinformatics,
statistics and cancer informatics, according to their interests
and relevance to the project theme. Apart from Dr. Campbell there
are 11 other academic researchers in the Intelligent Systems
Laboratory with interests in machine learning and bioinformatics,
including Nello Cristianini, Tijl de Bie and Julian Gough, in
addition to numerous research assistants and PhD students.

The Research Assistant will be hosted by the Intelligent Systems
Laboratory within the newly-formed Dirac School of Engineering,
Bristol University. Dr. Campbell’s publications are available at
http://www.enm.bris.ac.uk/cig/pubs.html and his personal webpage
at http://www.enm.bris.ac.uk/research/neural/staff/icgc.html.

The closing date for applications is:
***9.00am, Monday 1st March 2010***

Grade : Level A in Pathway 2
Salary : £29,853 – £33,600

Timescale of Appointment: fixed term contract for 2 years.

Further details and an application form can be found at:

https://www.bris.ac.uk/boris/jobs/ads?ID=85571

Alternatively you can telephone (0117) 954 6947,
minicom (0117) 928 8894 or email: Recruitment(at)bris.ac.uk,
quoting reference number 15239.

Summer School on Statistical Inference in Computational Biology

CALL FOR PARTICIPATION
SICSA International Summer School on
Statistical Inference in Computational Biology
National E-Science Centre, Edinburgh, United Kingdom
14-18 June 2010
http://www.dcs.gla.ac.uk/inference/sicb/

CONFIRMED SPEAKERS
—————-
Terry Speed (University of California, Berkeley)
Michael Stumpf (Imperial College)
Dirk Husmeier (Biomathematics and Statistics Scotland)
Chris Holmes (University of Oxford)
Manfred Opper (Technical University of Berlin)

SCOPE
—–

Technological advances in the life sciences are producing vast amounts of data describing organisms at all levels of organisation. The impact of this on Informatics and the Computational Sciences has been enormous: the new disciplines of computational biology and bioinformatics were born to organise and model these data, and are now some of the fastest growing and most exciting areas in computer science. The increasing awareness of the noisy and incomplete nature of most biological data has led to a widespread use of statistical and machine learning tools within the field.

The school will focus on the role of statistical inference in biological modelling, with a particular emphasis on the Bayesian framework. It is mostly aimed at PhD students in computational subjects or quantitative biology, although early career researchers wishing to acquire more statistical modelling skills are also welcome. The school will consist of 6 4-hour modules, each delivered by an expert of international standing over 5 days. The first two sessions will serve as an introduction to multi-variate and Bayesian statistics respectively with a leaning towards the tools required in Computational Biology. The remaining sessions will cover four of the main inference tasks in Computational Biology – network reconstruction, inference within models of biological processes, inference in phylogenetics and phenotype-genotype associations to explain genetic diseases.

For more information, email sicb(at)dcs.gla.ac.uk

Organisers: Guido Sanguinetti (University of Edinburgh) Simon Rogers (University of Glasgow)

Summer School on Cognitive Science and Machine Learning

We invite applications for the PASCAL2 Summer School on Cognitive Science and Machine Learning that will be held at Sardegna Ricerche (Italy) in May 2010.

http://www.mlss.cc/sardinia10

The summer school will take place immediately before and near the AISTATS 2010 conference.

=============================

Theme of the Summer School:

Cognitive science aims to reverse engineer human intelligence; machine learning provides one of our most powerful sources of insight into how machine intelligence is possible. Cognitive science therefore raises challenges for, and draws inspiration from, machine learning; and insights about the human mind may help inspire new directions for machine learning. This summer school brings together leading researchers from both fields, and those working at the interface between them. It is aimed at graduate students, post-docs and established researchers from both the cognitive science and machine learning communities, interested in exploring the interface between human and machine intelligence.

=============================

Confirmed Speakers (partial list):

Nick Chater, University College London
Alex Clark, Royal Holloway University of London
Silvia Chiappa, Cambridge University
Peter Dayan, University College London
Tom Griffiths, UC Berkeley
Konrad Körding, University of Chicago
Neil Lawrence, Manchester University
Bernhard Schölkopf, Max Planck Institute for Biological Cybernetics
Satinder Singh, University of Michigan
Josh Tenenbaum, MIT
Chris Watkins, Royal Holloway University of London
Felix Wichmann, Technical University of Berlin

=============================

Organisers:

Nick Chater, University College London
Silvia Chiappa, Cambridge University
Tom Griffiths, UC Berkeley
Neil Lawrence, Manchester University
Bernhard Schölkopf, Max Planck Institute for Biological Cybernetics
Josh Tenenbaum, MIT

=============================

Important Dates:

Application Submission Deadline: March 1 2010
Notification: March 26 2010
Subscription Deadline: April 1 2010
School: from May 6 to May 12 2010

Two Postdoc Positions: Vision group at Xerox Research Centre Europe

Xerox Research Centre Europe, based in Grenoble, France, is currently looking for two researchers with skills than can both integrate well and complement the current team; one focused on large-scale methods and the other on image analysis and feature extraction.

The Text and Visual Pattern Analysis Area (TVPA) group at Xerox Research Centre Europe, Grenoble, is a world-leading team that specializes in understanding, organizing, retrieving and enhancing both visual and hybrid content. Our research is the result of combining state of the art knowledge and skills from different fields such as machine learning, large-scale data mining, image analysis, and text retrieval. We have extensive experience and state of the art methods in image categorization, image enhancement, quality assessment and document image processing for both text and image content. Our main research lines currently focus on using this knowledge and expertise in the challenging areas of applied visual aesthetics and hybrid information access; going beyond the standard classification approach and developing techniques applicable in a variety of domains for assisted content creation and management.

Two candidates are sought with skills than can both integrate well and complement the current team; one focused on large-scale methods and the other on image analysis and feature extraction. This is an opportunity to join the group and research centre at a key time, and we are seeking researchers who will relish the challenge of not only carrying out leading research — through strong collaboration with Xerox researchers and also the wider academic community — but also influence the research agenda and potentially see strong deployment to be used worldwide in client systems.

For both positions, a strong background in machine learning and image or text processing is essential. Working with the team, the responsibilities will include inventing and developing novel techniques for document content analysis, both in terms of visual, text or hybrid media content. More specifically, there is currently a strong focus on large-scale learning and leveraging different types of media and social aspects to improve performance. The particular requirements for each position are detailed below:

Large-scale methods for retrieval and processing: the amount of digital content now stored and processed in a wide variety of application domains is every-increasing and the need for truly effective large scale techniques is clear. We are looking to develop retrieval and analysis methods that are computationally effective and can be used on image content, scanned forms, text-image hybrid data and even handwritten text. We are particularly seeking individuals with experience in one or more of large-scale methods, use of hybrid image/text data and handwriting recognition.

Image analysis and feature extraction: the team currently has state-of-the-art image processing techniques which have been repeatedly shown to be successful, through the strong publication record of the group, through excellent performance in competitions such as ImageCLEF and through deploying the technologies in various industrial applications. We are seeking a researcher with a blend of image analysis and machine learning skills that can use these technologies in different domains, and also develop the next generation of such methods. Experience with video analysis is desirable, as digital media of this type is increasingly common and is one of the focuses of Xerox customers for future digital asset management and real-time use of image processing technology.

Requirements (both positions):

* PhD in Computer Science in the area(s) of Machine Learning and/or Computer Vision

* Strong publication record and evidence of implementing systems

* Strong English-language written and oral communications skills

Informal enquiries are welcome and can be made in the first instance to the area manager: craig.saunders(at)xrce.xerox.com

To submit an application, please send your CV and cover letter to both xrce-candidates(at)xrce.xerox.com and craig.saunders(at)xrce.xerox.com.

About XRCE

More information about the position can be found at: http://www.xrce.xerox.com/About-XRCE/Jobs/Two-researchers-sought-for-leading-Textual-Visual-Pattern-Analysis-group

Workshop: Foundations and New Trends of PAC Bayesian Learning

============================================
Workshop:

Foundations and New Trends of PAC Bayesian Learning

University College London, UK

22 – 23 March 2010

http://www.cs.ucl.ac.uk/staff/rmartin/pacbayes/

CALL FOR PAPERS
Deadline: Friday, 12th February 2010
============================================

PAC-Bayes theory is a framework for deriving some of the tightest generalization bounds available. Many well established learning algorithms can be justified in the PAC-Bayes framework and even improved. PAC-Bayes bounds were originally applicable to classification, but over the last few years the theory has been extended to regression, density estimation, and problems with non iid data. The theory is well established within a small group of the statistical learning community, and has now matured to a level where it is relevant to a wider audience. The workshop will include tutorials on the foundations of the theory as well as recent findings through peer reviewed presentations.

Workshop topics

PAC Bayes theory or applications. In particular: application to:

* regression
* density estimation
* hypothesis testing
* structured density estimation
* non-iid data
* sequential data

The Invited Speakers include:

Olivier Catoni
CNRS U.M.R. 8553

David McAllester
Toyota Technological Institute at Chicago

Matthias Seeger
Saarland University and Max Planck Institute for Informatics

Organisers: Jean-Yves Audibert, Matthew Higgs, Steffen Grünewälder, François Laviolette and John Shawe-Taylor

Contact:
Steffen Grünewälder
steffen(at)cs.ucl.ac.uk

Call for papers: The Fifth European Workshop on Probabilistic Graphical Models (PGM’2010), Helsinki, Finland, September 13-15, 2010

PGM’2010 – FIRST CALL FOR PAPERS
================================

The Fifth European Workshop on Probabilistic Graphical Models (PGM’2010)
Helsinki, Finland, September 13-15, 2010
http://www.helsinki.fi/pgm2010/

Key dates:
==========
* Deadline for submissions: June 4, 2010
* Notification of acceptance: July 16, 2010
* Final versions due: July 30, 2010
* Workshop: September 13-15, 2010

Call for papers:
================

The European Workshop on Probabilistic Graphical Models (PGM) is a biennial workshop, which was first held in Cuenca, Spain, in 2002, followed by workshops in Leiden (2004), Prague (2006) and Hirtshals (2008). The fifth PGM workshop will be held in Helsinki, Finland, September 13-15, 2010.

The aim of the workshop is to bring together people interested in probabilistic graphical models and provide a forum for discussion of the latest research developments in this field. The workshop is organized so as to facilitate discussions and collaboration among the participants also outside the workshop sessions. We welcome theoretical and applied contributions related to the following topics:
* Principles of Bayesian (belief) networks, chain graphs, decision networks, influence diagrams, and other probabilistic graphical models (PGMs)
* Information processing in PGMs, exact and approximate inference
* Learning in the context of PGMs: machine learning approaches, statistical methods, data/graph mining, criteria for model selection/validation/regularization, optimization algorithms for searching the model space
* Exploitation of results from different disciplines for the construction and use of PGMs, e.g., computer science, statistics, information theory, mathematics, physics, optimization, decision theory
* Software systems based on PGMs
* Applications of PGMs to real-world problems

We invite submissions that concern one or more of the above topics, or any other aspects related to probabilistic graphical models. Papers submitted for review should report on original, previously unpublished work. Each submitted paper will be reviewed by at least two reviewers.

Submission procedure: see http://pgm2010.hiit.fi/cfp.html

Programme Co-Chairs:
Petri Myllymäki, University of Helsinki
Teemu Roos, Helsinki Institute for Information Technology HIIT
Tommi Jaakkola, MIT

Contact: pgm2010(at)helsinki.fi
Home page: http://www.helsinki.fi/pgm2010/