Post doctoral position in Machine Learning/Cognitive Vision/CBIR

http://institute.unileoben.ac.at/infotech/research/jobs/nfn-position-2009-06.html

For a project funded by the Austrian Science Foundation (FWF) and the European Commission, we are looking for a highly motivated post doctoral researcher with background in machine learning and/or cognitive vision and/or content based image retrieval. Among the possible fields of specialization are on-line learning, active learning, reinforcement learning, visual object classification, relevance feedback, and query-by-example search.

To learn more about the above project and the research at the Chair of Information Technology, University of Leoben, Austria, please visit http://institute.unileoben.ac.at/infotech.

This position will be filled as soon as possible for the duration of 12 month (with a possible extension). Depending on your qualification salary is 30000-45000 EUR per year (after paying all social and insurance benefits and taxes this is net 1500-2000 EUR per month). Highly qualified PhD candidates may be considered as well.

Applicants should submit 1) a CV, including a brief research statement, 2) 1-3 recent publications in electronic format, and 3) the names and contact information of three individuals who can serve as references.

Contact:
Univ.-Prof. Dr. Peter Auer
University of Leoben
Chair for Information Technology
Franz-Josef-Strasse 18, A-8700 Leoben, Austria
Fax: +43(3842)402-1502
E-mail: auer (at) unileoben.ac.at

PASCAL2 Workshop on Spatiotemporal Modelling

http://homepages.inf.ed.ac.uk/mdewar1/workshops/spatiotemporal/

12th – 14th October 2009
Informatics Forum, Edinburgh, UK

Submission Deadline: July 30th 2009

CALL FOR PAPERS

Recent years have witnessed a considerable amount of research on modelling spatio-temporal systems both in an engineering and scientific domain. The combination of novel theoretical problems, important applications areas and mix of disciplines in spatiotemporal modelling offers real potential for some intellectual excitement.

The workshop aims to bring together the various interested communities (machine learning, control, statistics, geosciences, econometrics, ecology etc.), to focus on theoretical aspects of spatiotemporal modelling and to encourage cross pollination of ideas.

The workshop will be a two-day , single-track workshop, with two half days and one full day in between. We have 6 invited speakers and would like to encourage submission of abstracts for additional talks and posters. The oral sessions will be relatively short, punctuated by breaks and poster sessions to encourage discussion. We hope that submissions will come from a range of topics and applications surrounding spatiotemporal modelling, for example:
* learning and inference
* multi-scale modelling
* experiment design
* spatiotemporal covariance functions
* heterogeneous models
* sensor networks
* disease mapping
* …

KEY SPEAKERS

* Manfred Opper, AI, Berlin.
* Sujit Sahu, Mathematics, Southampton.
* John Haslett, Statistics, Trinity College Dublin.
* Chris Glasbey, Biomathematics & Statistics Scotland, Edinburgh.
* Geir Storvik, Mathematics, Oslo.
* (and hopefully) Dan Cornford, Computer Science, Aston.

IMPORTANT DATES

* Submission Deadline: August 14th 2009
* Author Notification: August 31st 2009
* Workshop: 12th – 14th October 2009

PAPER SUBMISSION

We welcome submissions of one page abstracts, in the Springer LNCS style. You can submit abstracts here:
http://homepages.inf.ed.ac.uk/mdewar1/workshops/spatiotemporal/submission.html

ORGANISERS

* Dr Michael Dewar
* Prof Chris Glasbey
* Prof Chris Williams
* Dr Amos Storkey
* Dr Guido Sanguinetti
* Prof Visakan Kadirkamanathan

Funded PhD position in machine learning at the University of Technology of Compiègne (France)

The Laboratory Heudiasyc, CNRS & University of Technology of Compiègne (France) invites applications for a fully funded 3-years PhD position.

Topic: “Model Selection in Block-Clustering”

Starting date: October 2009

Block clustering aims at partitioning simultaneously the rows and columns of a data matrix. This problem, which has been investigated from the early 80’s in data analysis, has recently attracted interest in the machine learning community, thanks to applications such that text mining, marketing or recommender systems (see e.g. Netflix challenge). This thesis aims at developing new model selection tools for block clustering, starting from their theoretical foundations, up to their empirical evaluation. It will be developed in the ClasSel project funded by the French National Research Agency.

Clustering, and particularly block-clustering is highly contingent on the choices of the fitting criterion and the number of classes. Viewing the clustering problem as a density estimation problem provides answers to these problems thanks to the general-purpose estimation and model selection tools developed in the statistical framework. In this context, model selection is usually performed via penalized maximum likelihood scores, that trade-off fitting for model complexity. The main approaches are AIC (Akaike Information criterion) that builds on Kullback-Leibler divergence and BIC (Bayesian Information criterion) that maximizes the model posterior. However, these criteria do not take into account the classification objectives when density models are used in the clustering framework. A third criterion, ICL (Integrated Completed Likelihood) accounts for this aspect. Finally, resampling methods, though scarcely used in clustering, may also provide estimates of the optimism of complex models.

All these criteria are sample-size dependent. Hence, their adaptation to block-clustering is problematic since problem size is characterized here by the number of rows and the number of columns of the data matrix. The notion of sample size cannot be transposed directly, and the first attempts to apply information criteria in this framework are not convincing. This project aims at proposing such criteria, either by revising the standard criterion, or by starting from new theoretical grounds.

Applicants should complete their Master degree in 2009, and have interest and expertise in machine learning, statistics, computer science and/or applied mathematics.

Expressions of interest with a short CV, including reference(s) should be sent to: Gerard.Govaert (at) utc.fr or Yves.Grandvalet (at) utc.fr

Some links:
– Heudiasyc Lab. http://www2.hds.utc.fr/index.php?id=1&L=1
– University of Technology of Compiègne http://www.utc.fr/the_university/index.php
– Compiègne is a small town with 70’000 inhabitants, at a 45mn commute of Paris, see http://en.wikipedia.org/wiki/Compi%C3%A8gne

Deadline Extension: Robotics workshop at ECML PKDD

The deadline for sending papers has now been extended to 15 June 2009.

Learning and Data Mining for Robotics: Call for Papers
Workshop collocated with ECML-PKDD 2009, Sept. 7th, Bled, Slovenia
http://www.i.kyushu-u.ac.jp/~suzuki/lemir.html

Why … What … When … Who…

WHY: Designing robot controllers for an open world is among the grandest challenges relevant to Machine Learning and Cognitive Systems at large. The goal of the workshop is to pro-actively discuss the research agenda of Learning and Mining for Robotics: to agree on what can be considered as solved problems, to list the current scientific priorities, to propose milestones, to share ideas, know-how and links with actual robotic platforms.

WHAT: Among the most critical tasks are the modeling of the robot itself (sensor/actuators), the modeling of the robot world (including other robots), the modeling of the task (reinforcement, imitation/apprenticeship, control), the design of experiments and active learning (non exclusive list, NEL). All contributions related to Bayesian and probabilistic robotics, reinforcement learning and apprenticeship learning, learning from sensor data and logs (NEL) are welcome. All contributions discussing issues such as: how to learn with/without a simulator, and how to face the reality gap; how to build controllers that are robust with respect to motor/sensor failures; how to derive self-driven criteria/goals for the robot; how to deal with swarm robotics, distributed decision making and frugal communication; how to endow the robot with an implicit memory; are equally welcome. Finally, all contributions related to the exploitation of the robotic logs in order to debug/assess/reverse engineer the controller are welcome.

WHEN and HOW: Deadline: June 15th
Send your research paper/position paper (Springer Verlag, < 10 pages) via http://www.i.kyushu-u.ac.jp/~suzuki/lemir.html Notification: June 30th Camera ready: Aug. 15th Workshop: Sept. 7th Style: http://www.springer.com/computer/lncs/lncs+authors SGWID=0-40209-0-0-0 WHO: Chairs: Einoshin Suzuki, Kyushu Univ. and Michele Sebag, CNRS PC: Shin Ando, Gunma Univ. Jose L. Balcazar, TU Catalonia Aude Billard, EPFL Nicolas Bredeche, Univ. Paris-Sud Joao Gama, Univ. Porto Peter Grunwald, CWI Hitoshi Iba, Univ. of Tokyo Kristian Kersting, Fraunhofer IAIS Jan Peters, Max Planck Institute Tubingen Marc Schoenauer, INRIA Marc Toussaint, TU Berlin Takashi Washio, Osaka Univ. Workshop Web page: http://www.i.kyushu-u.ac.jp/~suzuki/lemir.html

Research Fellow in Computer Vision and Machine Learning

Reference: 076/16878/AAP

School of Technology, Oxford Brookes
Research Fellow in Computer Vision and Machine Learning
Starting salary: £27,998, rising annually to £30,595

The objective of this proposed project is to develop a high performance human pose and motion capture and analysis (HPMCA) proof of concept demonstrator, based on novel technologies protected by copyright and a patent application, which have been developed in the Computer Vision Group at Oxford Brookes University. This demonstrator, which will not be dependent on additional hardware, such as motion capture devices or specialised cameras, will be the basis for further commercialisation activities, which will include licensing, service provision and attracting venture funding for a start-up company to develop and exploit the technologies across a range of different market sectors.

More information available here.

Call for Papers ECML-PKDD 09 WS: Learning from non-IID Data

Workshop on Learning from non-IID data: Theory, Algorithms and Practice
During ECML-PKDD 2009
7 September 2009, Bled, Slovenia

Last Call for Papers
Please note the change of the submission deadline, that is now June 14th, 2009.
Note that we now expect interested researchers to submit abstracts (200/300 words) by June 11th, 2009.

DESCRIPTION
———–
Both classification and regression frameworks in Machine Learning were developed under the independently and identically distributed (IID) assumption. Though this assumption helps to study the properties of learning procedures (e.g. generalization ability), and also guides the building of new algorithms, there are many real world situations where it does not hold. This is particularly the case for many challenging tasks of machine learning that have recently received much attention such as (but not limited to): ranking, active learning, hypothesis testing, learning with graphical models, prediction on graphs, mining (social) networks, multimedia or language processing.

In this workshop, we will discuss recent developments, explore new issues and share ideas for future directions for learning from non-IID data. We welcome papers that address any of the above questions or that focus on any of the following topics:

* Theoretical: results on generalization bounds and learnability, contributions that mathematically formalize the types of non-IIDness encountered, results on the extent to which non-IIDness does not harm the validity of theoretical results built on the IID assumption, helpfulness of the online learning framework,
* Algorithmic: theoretically motivated algorithms designed to handle non-IID data, approaches that make it possible for classical learning results to carry over, online learning procedures,
* Practical: successful applications of non-IID learning methods to learning from streaming data, web data, biological data, multimedia, natural language, social network mining.

SUBMISSION
———-
Please send to lniid09@liste.lif.univ-mrs.fr by email the following:

– A full paper up to 8 pages for one of the 3 tracks:
(1) Oral presentation,
(2) Poster spotlights,
(3) Posters.

– In the body of your email, include (in plain ASCII): names of all authors, their affiliations, their physical and email addresses and the track number which corresponds to your submission.

Submissions will be reviewed for technical soundness, relevance, significance and clarity by the organizing and review committee and invitations to present will be sent accordingly.

The full paper should be formatted according to the standard LNCS templates available at: http://www.springer.com/computer/lncs/lncs+authors?SGWID=0-40209-0-0-0 and submitted as a PDF or postscript file.

IMPORTANT DATES
—————
Abstract submission deadline: June 10, 2009
Paper submission deadline has been extended to: June 14, 2009
Notification of acceptance: June 30, 2009
Final camera ready submissions: August 15, 2009
Workshop: September 7, 2009

ORGANIZERS
———-
Massih-Reza Amini, National Research Council, Canada
Amaury Habrard, University of Marseille, France
Liva Ralaivola, University of Marseille, France
Nicolas Usunier, University Pierre et Marie Curie, France

PROGRAM COMMITTEE
—————–
Shai Ben-David, University of Waterloo, Canada
Gilles Blanchard, Fraunhofer FIRST (IDA), Germany
Stéphan Clémençon, Télécom ParisTech, France
François Denis, University de Provence, France
Claudio Gentile, University dell’Insubria, Italy
Balaji Krishnapuram, Siemens Medical Solutions, USA
François Laviolette, Université Laval, Canada
Xuejun Liao, Duke University, USA
Richard Nock, University Antilles-Guyane, France
Daniil Ryabko, INRIA, France
Marc Sebban, University of Saint-Etienne, France
Ingo Steinwart, Los Alamos National Labs, USA
Masashi Sugiyama, Tokyo Institute of Technology, Japan
Nicolas Vayatis, Ecole Normale Supérieure de Cachan, France
Zhi-Hua Zhou, Nanjing University, China

KEYNOTE SPEAKERS
—————-
Shai Ben-David, University of Waterloo, Canada
Nicolas Vayatis, Ecole Normale Supérieure de Cachan, France

—————————————————-
See the workshop Web page at:
http://www-connex.lip6.fr/~amini/ecml-wk-lniid.html

Regression in Robotics Workshop: Call for Participation

RSS’09/PASCAL2 Workshop on Regression in Robotics — Approaches and Applications

Sunday, June 28, 2009, Seattle, WA, USA

Co-located with Robotics: Science & Systems, RSS 2009
Sponsored by PASCAL2 Network of Excellence

http://www.robreg.org

Dear colleagues,

We invite you to attend this full-day workshop, to be held on Sunday, June 28, 2009 in Seattle at the University of Washington campus. The workshop will feature invited speakers, selected poster presentations and a moderated panel discussion.

Please note the following recent updates:

– We are pleased to announce that a Pascal Best Poster Presentation Award of US$ 350 will be given to the authors of a selected poster. Results will be determined by a panel of judges *and* by popular vote.

– After the workshop, we invite all participants to an informal Robot Learning Cocktail Night organized jointly with the RSS workshop on “Bridging the gap between high-level discrete representations and low-level continuous behaviors”.

Please refer to http://www.robreg.org for a more detailed program schedule.

Invited Speakers:

** Pieter Abbeel, University of California at Berkeley
** Dieter Fox, University of Washington
** Raia Hadsell, Carnegie Mellon University
** Andreas Krause, California Institute of Technology
** Jan Peters, Max-Planck Institute of Biological Cybernetics
** Rajesh Rao, University of Washington
** Nick Roy, Massachusetts Institute of Technology

Description:

Function approximation from noisy data is a central task in robot learning. Relevant problems include sensor modeling, manipulation, control, and many others. A large number of regression methods have been proposed from statistics, machine learning and control system theory to address robotics-related issues such as online updates, active sampling, high dimensionality, non-homogeneous noise and missing features. However, with minimal communication and collaboration between communities, work is sometimes reproduced or re-discovered, making research progress challenging.

Our goal is to draw researchers from the different communities of robotics, control systems theory and machine learning into a discussion of the relevant problems in function approximation to be learned in robotics. We would like to develop a common understanding of the benefits and drawbacks of different regression approaches and to derive practical guidelines for selecting a suitable approach to a given problem. In addition, we would like to discuss two key points of criticism in current robot learning research. First, data-driven machine learning methods do, in fact, not necessarily outperform models designed by human experts and we would like to explore what regression problems in robotics really have to be learned. Second, regression methods are typically evaluated using different metrics and data sets, making standardized comparisons challenging.

Goal & Topics:

The workshop will address topics such as the following:

*** Approaches: Which learning approaches have been applied successfully to solve regression problems in robotics or have a high potential for doing so?

*** Problem settings: Which robot learning problems contain regression or function approximation as a central component? What are the specific aspects that make the problem challenging?

*** Theoretical foundations: How can challenging requirements such as online updates, active sampling, high dimensionality, non-homogeneous noise and missing features be addressed?

*** Benchmarking and evaluation: What are suitable methods for evaluation of regression methods? What metrics are being used and, subsequently, which should be used? Which benchmark data sets are available and which are missing?

Workshop Organizers:

Christian Plagemann
Stanford University
plagemann (at) stanford.edu

Jo-Anne Ting
University of Edinburgh
jting (at) ed.ac.uk

Sethu Vijayakumar
University of Edinburgh
sethu.vijayakumar (at) ed.ac.uk

Summer School on on Advanced Technologies for a Knowledge-Powered Enterprise

“First ACTIVE Summer School on Advanced Technologies for a Knowledge-Powered Enterprise”
http://active09.ijs.si/
Bled, Slovenia, 4-6th September 2009

!!! limited participation to 40 !!!

You are cordially invited to attend the “1st ACTIVE Summer School” that will take place from September 4 – 6 in Park Hotel, Bled, Slovenia. The summer school precedes the European Conference on Machine Learning “European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases – ECML PKDD 2009”.

Organised by ACTIVE FP7 Integrated Project (http://www.active-project.eu) and hosted by the Jozef Stefan Institute, the summer school seeks to bring together students, scholars and researchers from industry being being in order to share the recent developments, solutions and technologies from the areas of::

*Semantic technologies and content,
*Social software and Web 2.0,
*Adaptive and context-aware systems,
*Context mining,
*Process mining,
*Knowledge filters,
*Stream mining,
*Anomaly detection,
*Meta learning,
*Forecasting, and
*Social network analysis.

During the three day summer school attendants will have an opportunity to listen to very distinguished invited speakers: Paul Warren (BT), Marko Grobelnik (JSI), Tobias Bürger (STI), Igor Dolinšek (HERMES SoftLab), Rayid Ghani (Accenture), Michael Witbrock (Cycorp), Lise Getoor (University of Maryland), Neel Sundaresan (eBay), Pat Moore (Bloomberg), Paolo Paganelli
(Insiel), Denny Vrandečić (KIT), Marcel Tilly (Microsoft), Ian Mulvany (Nature).

Participation is limited to 40 participants. Based on already expressed interest, we expect to receive more applications than there are places available. This is why we ask potential participants to send their expression of interest together with curriculum vitae in pdf format no later than July 1, 2009 to tina.anzic (at) ijs.si. Selected participants will be notified by July 15, 2009.

The registration fee including accommodation for three nights in a three star hotel and sharing a double room is only 120 EUR. Without accommodation the cost is 50 EUR (teaching materials, refreshments, lunches, and one social event included).

Detailed information about the school, admission and fees can be found at:
http://active09.ijs.si/

Questions about the “First ACTIVE Summer School” may be directed to
tina.anzic (at) ijs.si.

ACTIVE Learning Community

Sandpit on the Application of Machine Learning Techniques to the Analysis of Complex Biomedical Data

Joint National Institute of Medical Research UK, University College London, PASCAL Workshop:

Sandpit on the Application of Machine Learning Techniques to the Analysis of Complex Biomedical Data
6-7th July, 2009
National Institute for Medical Research,
London

URL: http://www.davidroihardoon.com/SCBD09/

Organisers: Delmiro Fernandez-Reyes, James Briscoe, Gunnar Raetsch, John Shawe-Taylor, David R. Hardoon

This workshop will bring together a group of researchers from the biomedical sciences, machine learning and computational biology, to address the solution of challenging problems and help to develop an edge over existing approaches. The workshop will be organized around a small set of biomedical researchers, who will describe datasets and associated analysis challenges. At the same time there will be an opportunity for machine learning researchers to give overviews of approaches that might be effective for these problems.

At the least we hope that the interchange will provide valuable insights for researchers from both communities into the problems and techniques. The ideal outcome would be the establishment of new cross-disciplinary collaborations and/or new challenges based on the biomedical data.

The workshop will be grouped around four themes each occupying half a day.

Theme 1: Sequence Analysis, Motifs and Signal Detection
Main presenters:

* Sebastien Gagneux, “Next-generation sequencing for the population genomics and epidemiology of the tubercle bacilli”
* James Briscoe TBA

Theme 2: Gene and Protein Expression Analysis
Main presenters:

* Diogo Castro, “Molecular Neurobiology Identifying the genome-wide targets of transcription factor Mash1”
* Ben Martynoga, “Molecular Neurobiology Genome-wide location of transcription factors that share common targets in neural stem cells”
* Robert Wilkinson, “Bioinformatic and empirical analysis of novel hypoxia-induced antigenic targets in M. tuberculosis”
* Anne O’Garra TBA

Theme 3: Systems Biology and Medicine
Main presenters:

* Ben Seddon, “Immune Cell Biology Genetic networks regulated by IL7 in T cells”
* Thea Hogan, “Immune Cell Biology Modelling proliferation, survival and differentiation in T cell homeostasis”
* Delmiro Fernandez-Reyes TBA

Theme 4: Bio-imaging

* Peter Rosenthal, “Computing Cellular Architecture Systems Biology”

We welcome contributions of short presentations of approximately in areas of Machine learning addressing the four themes in order to promote the interaction. These could include but are not limited to:

* Statistical inference in High Dimensional Spaces
* Unsupervised, Semi-Supervised and Supervised Learning
* Variable and Feature Selection
* Clustering
* Structured output learning
* Graph and manifold learning

Please submit your proposed contribution to email sandpit@cs.ucl.ac.uk before June 21.

Postdoc position at Xerox

Position: postdoc researcher, Cross-Language Technologies

The Cross-Language Technologies (CLT) group of the Xerox Research Centre Europe has the mission of researching and developing the most effective methods for crossing language barriers. Targeted applications include Statistical Machine Translation, multilingual terminology extraction, Cross-Language Information Retrieval (CLIR), Categorization and Clustering. The team is part of the PASCAL 2 European Network of Excellence, ensuring a strong network of academic collaboration.

We are looking for a postdoc researcher to contribute to our Statistical Machine Translation and CLIR activities in the framework of an EU funded project.

XRCE is a highly innovative place and we strongly encourage publication and interaction with the scientific community.

Required experience and qualifications:

* PhD in computer science, statistics or mathematics with excellent knowledge of machine translation, cross-language information retrieval or machine learning for natural language processing
* Good programming skills in C, C++, and/or Python.
* A good command of English is required, as well as open-mindedness and the will to collaborate with a team.
* Experience in machine learning is a definite plus.

To apply: Please email your CV and covering letter, with message subject “Cross-Language Technologies Researcher” to xrce-candidates at xrce.xerox.com and to Nicola.Cancedda at xrce.xerox.com.

Inquiries can be directed to Nicola.Cancedda at xrce.xerox.com.

The duration of the contract is 18 months, starting around September 1st, 2009.

Xerox Research Centre Europe (XRCE) is a young, dynamic research organization, which creates innovative new business opportunities for Xerox in the digital and Internet markets. XRCE is a multicultural and multidisciplinary organization set in Grenoble, France. We have renowned expertise in machine learning, work practice studies, image processing, natural language processing and document structure. The variety of both cultures and disciplines at XRCE makes it both an interesting and stimulating environment to work in, leading to often unexpected discoveries! XRCE is part of the Xerox Innovation Group made up of 550 researchers and engineers in four world-renowned research and technology centres. The Grenoble site is set in a park in the heart of the French Alps in a stunning location only a few kilometers from the city centre. The city of Grenoble has a large scientific community made up of national research institutes (CNRS, Universities, INRIA, Minatec) and private industries. Grenoble is close to both the Swiss and Italian borders and is the ideal place for skiing, climbing, hang gliding and all types of mountain sports.