News Archives

PostDoc on road traffic datamining at Mines ParisTech

Applications are invited for a post-doctoral position in road traffic datamining and prediction, for 15 month starting within S1 2009, at Robotics Lab. of Mines ParisTech (Paris, France).

The robotics laboratory (CAOR) of Mines ParisTech, associated with IMARA project of INRIA in LaRA « Joint Research Unit », has been involved in 2 big European projects (REACT and COM2REACT) using V2I (Vehicle toInfrastructure) and V2V (Vehicle To Vehicle) communications for enhancing global « road information system ». In these projects the 2 labs worked in particular on analysis and prediction of traffic for improving preventive re-routing strategies in order to reduce congestions. An algorithm has been developed for reconstructing and predicting traffic from a fleet of « sensor » vehicles regularly sending position/speed/traffic information.
CAOR has just started, again with IMARA, and together with TAO project of INRIA and LET of Lyon, a new collaborative project sponsored by ANR (French national research funding agency). This project will focus on analysis and prediction of road traffic, first on realistic simulated data (to be produced with Metropolis software developed by LET), then on real data.

Research work description
The work to do is firstly data-mining of traffic data, seen as a graph whose each edge is a road section with associated traffic level (mean speed, travel time or congestion level), in order to extract common traffic patterns. For this « pattern mining », the idea is to test various clustering methods, in particular unsupervised training algorithms, such as Kohonen maps and K-means, so as to identify main « attractor » states and/or usual traffic states.
Then, the candidate should try to build a simplified dynamic model, as prediction of transitions between identified patterns. In particular, the possibility to exploit fully the graph structure of roads network shall be examined, by experimenting “graph kernel methods” recently developed and mainly applied in the context of bioinformatics.
A possible extension is analysis of road network as a complex dynamical system (bifurcation diagram, etc…).

The candidate should hold a good PhD in the field of statistical machine-learning and/or data-mining, with:
• Very good knowledge of data mining and analysis techniques, as well as of machine-learning methods;
• Good knowledge in probabilities and statistics (in particular Markovian models);
• Some knowledge on graphs and associated algorithms;
• Good computer programming skills (C/C++/Java)

Speaking French is not absolutely mandatory, but would be a plus.

Duration and date
Duration of post-doctoral contract is 15 month, starting within first semester 2009.

Supervision and contact :
Fabien Moutarde, (+33), Fabien.Moutarde (at)

To apply:
Candidates must send a detailed CV, with a cover letter, main publications (or links), together with name and contact of at least 2 references, to above e-mail address.

The Analysis of Patterns – Call for Participation

You are invited to participate in the third course on:

Pula Science Park (Cagliari)
Pula, Italy
September 27th – October 3rd, 2009

Organizers: Nello Cristianini, Fabio Roli, Tijl de Bie


* Florent Nicart – Université de Rouen
* Jean Philippe Vert – Mines Paris Tech
* John Shawe-Taylor – University College London
* Nello Cristianini – University of Bristol
* Bart Goethals – University of Antwerp
* Elisa Ricci, Idiap
* Fabio Roli, University of Cagliari

+ Research Seminars (to be announced)


Every aspect of modern society has been affected by the data revolution. Cheap collection, storage and transmission of vast amounts of information have revolutionized the practice of science, technology and business. Ideas from various disciplines have been deployed to help in the task
of designing computer systems that can automatically detect and exploit useful regularities (patterns) in general types of data.

This is the third meeting of a series devoted to pursuing a unified theoretical description of the various branches of Pattern Analysis. These include statistical approaches to pattern recognition,
combinatorial approaches to pattern matching, grammatical representations of structures, and many more fields of mathematics and computer science. The summer school will aim to emphasize a fundamental unity in goals and methods in all these diverse fields, to enhance our understanding of the central principles of pattern analysis, and to assist in the development of new pattern analysis approaches.

The meeting is interdisciplinary in nature, and can be seen both as a School for advanced students, and as a Workshop for researchers. Leading researchers in various subfields of pattern analysis will hold tutorials on their subject area, while new ideas will be presented in poster sessions, discussions and short seminars. Students in machine learning, pattern recognition, statistics, optimization, data mining, bioinformatics, are particularly encouraged to apply.

The registration fee for the School is 680 Euro per person for a double room (820 Euro for a single room) and includes 7 nights accommodation, meals and school fees.

Attendance is limited to 50 students and will be allocated on a first-come-first-served basis.

Registrations will open in february 2009.

Final Call for ICML/UAI/COLT 2009 Workshop Proposals

Montreal, Canada, June 18 2009
Proposal Deadline: Mon 19 Jan, 2009
Acceptance Notification: February 2, 2009

The ICML, UAI, and COLT conferences will be colocated in Montreal June
14-21 2009. We solict proposals for workshops to be held during a single joint workshop day on June 18. This date lies between ICML (June 14-17) and UAI/COLT (June 19-21). Workshops will be selected on the basis of their interest to the attendees of one or more of the conferences.

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 encouraged. Representation of alternative viewpoints and panel-style discussions are also encouraged.

* Organization

The format, style, and content of accepted workshops is under the control of the workshop organizers and largely autonomous from the main conferences. The workshops will be seven hours long and split into morning and afternoon sessions. Workshop organizers will be expected to manage the workshop content, specify the workshop format, be present to moderate the discussion and panels, invite experts in the domain, and maintain a website for the workshop. Workshop registration will be handled centrally by the main conferences with a single uniform registration fee and with registrants allowed to attend workshops other than the one they register for.

* Submission Instructions

Proposals should specify clearly all of the following:

* the workshop’s title (what is it called?)
* topic (what is it about?)
* motivation (why a workshop on this topic?)
* impact and expected outcomes (what will having the workshop do?)
* potential invited speakers (who might come?)
* a list of related publications (where can we learn more?)
* main workshop organizer (who is making it happen?)
* other organizers (who else is making it happen?)
* workshop URL (where will interested parties get more information?)
* relevant conferences (which of ICML, UAI, and COLT would it appeal to?)

Please also provide brief CVs of all organizers.
This information should be sent by email (in plain text or pdf format) to
Icml-uai-colt-workshops09 (at)
by 19 Jan 2009.

Jeff Bilmes and Andrew Ng: UAI co-chairs Sham Kakade: COLT workshops chair Chris Williams: ICML 2009 workshops chair

Call for papers: Journal of Machine Learning Research , Special Topic on Large Scale Learning – Dedline Extension

With the exceptional increase in computing power, storage capacity and network bandwidth of the past decades, ever growing datasets are collected in fields such as bioinformatics (Splice Sites, Gene Boundaries, etc), IT-security (Network traffic) or Text-Classification (Spam vs. Non-Spam), to name but a few. While the data size growth leaves computational methods as the only viable way of dealing with data, it poses new challenges; specifically, most machine learning algorithms hardly scale up beyond 1,000,000 examples or dimensions.

A special topic of the Journal of Machine Learning Research will be devoted to Large Scale Learning, in the line of the NIPS 2007 and ICML 2008 “Efficient Machine Learning” Workshops, and of the Pascal Challenge on Large Scale Learning (

You are invited to submit your contributions to this special issue. For the sake of a principled and fair evaluation, binary classification algorithms must be assessed on the datasets and along the experimental protocol devised for the Large Scale Learning Challenge. More information about the challenge protocol can be found here:

Important dates

Submission: 5 February 2009 ***NEW***
Decision: 15 March 2009
Final versions: 15 April 2009

Topics of Interest

Topics of interest include:

* Applications to very large scale problems in, e.g., bioinformatics, textcategorization, network data
* Efficient training algorithms, e.g., SVMs solvers
* Learning with a budget, e.g., under strict time or memory constraints.
* Efficient parallelization of machine learning algorithms
* Efficient data structures
* On-line learning algorithms
* Large-scale kernel methods
* Coarse to fine algorithms
* Algorithms making use of new hardware, e.g., GPUs, Xilinx

Submission procedure

Authors are kindly invited to follow the standard JMLR format and submission procedure JMLR submission format, the number of pages is limited to 30. Please include a note stating that your submission is for the special topic on Large Scale Learning.

Guest editors

Soeren Sonnenburg, Fraunhofer Institute FIRST, Berlin, Germany
Vojtech Franc, Fraunhofer Institute FIRST, Berlin, Germany
Elad Yom-Tov, IBM Haifa Research Lab, Haifa, Israel
Michele Sebag, LRI, Orsay, France

Fully Funded PhD Studentship in Systems Biology

British Heart Foundation Fully Funded PhD Studentship in Systems Biology


Dr. Tim Palmer (, Integrative and Systems Biology)
Prof. Mark Girolami (,

Regulation of the immune/inflammatory responses by interleukin-6 (IL-6)-family cytokines is dictated by the interplay of multiple cytokine-activated signalling cascades and inhibitory regulators designed to prevent excessive receptor activation that can result in disease. The situation is further complicated by the observation that cytokine-activated signalling cascades are negatively controlled by distinct signalling modules such as those initiated by the prototypical intracellular messenger cyclic AMP. Despite its significance, the extensive level of cross-talk observed has not been integrated into coherent models of IL-6 receptor signalling and its regulation.

By combining molecular/cell biology with mathematical modelling & statistical inferential approaches, this inter-disciplinary studentship will A) statistically define minimal network structures that accurately describe cytokine signalling pathway kinetics, B) derive a set of plausible mathematical models that can identify the critical parameters controlling inhibitory cross-regulation of gp130 by cyclic AMP, and C) identify new approaches for limiting excessive cytokine signalling associated with inflammatory disorders.

The project provides an exciting opportunity for high-quality doctoral training in mathematical modelling & statistical inferential approaches and their application to increase our understanding of the architecture and dynamics of molecular cell signalling pathways. In addition to contemporary
molecular and cellular biology techniques (mammalian cell culture, RNAi-mediated knockdown, protein analysis), the successful candidate will be trained in mathematical modelling of pathway dynamics as well as Bayesian statistical methods to formally characterize uncertainty in these models.

Candidates should be European Economic Area nationals, have an excellent first degree in a relevant mathematical discipline (Mathematics, Computing Science, Statistics, Engineering, Physics) and be highly motivated in their wish to apply this expertise to biological systems. Candidates with an excellent first degree in Biochemistry, Molecular and Cellular Biology or a related discipline, coupled with additional experience in applying mathematical/statistical methods to biological systems, will also be considered.

The studentship will commence as soon as possible after a suitable candidate is identified. The studentship will carry a stipend of £16,853 in year 1 and increasing to £18,580 in year 3. The studentship also covers the student¹s university fees. The studentship is renewable, subject to
satisfactory annual progress, for up to a total of three years.

Applications must consist of a current CV, contact details of at least two academic referees, evidence of degree performance, and a completed application form from

Preliminary email enquiries to Tim Palmer or Mark Girolami are welcomed.

Candidates are encouraged to complete the online application, but also to send their CV and associated documents direct to the Graduate School:

Graduate School of Biomedical and Life Sciences, Bower Building,
University of Glasgow, Glasgow G12 8QQ
Tel: ++44 (0)141-330-5800
Fax: ++44 (0)141-330-6093
E-mail: biograd (a) (please type ³BHF Palmer² in the subject box of

Post doctoral position in Machine Learning 2009

Starting: March 2009
Location: Paris, France

TOPICS: machine learning for structured data, social networks, random graphs, graphical models

The Department Signal and Image Processing (TSI) of Telecom ParisTech (France) is offering a one year post-doctoral position in Machine Learning. The post-doctoral fellow will develop and implement machine learning procedures and statistical techniques for investigating the diffusion of information through small social networks. The context of the study is related to food safety and dietary risks.

Candidates will be recruited at the level of a PhD in Mathematics or Statistics. They will have confirmed skills in mathematical modelling, data analysis, statistical or machine learning methods, mathematical programming (Matlab or R), and will be highly motivated for applications to social sciences.

The candidate will enjoy a challenging and rewarding working environment, within a top leading laboratory in the field of Information and Communication Theory.
Team members: Stéphan Clémençon (Telecom ParisTech – TSI), Fabrice Rossi (Telecom ParisTech – INFRES), Nicolas Vayatis (ENS Cachan – CMLA), Sandrine Blanchemanche (INRA Unité Met@risk), Akos Rona-Tas (UCSD Dept of Sociology).

INSTITUTION: Department of Signal and Image Processing of Institut Telecom – and Laboratory LTCI UMR Telecom ParisTech/CNRS 5141 –

FUNDING: position is funded by a new grant « Futur & Rupture « (Institut Telecom).

NET SALARY: ranging from 2200 to 2700 euros per month depending on past experience

CONTACT: Interested applicants should sent C.V. to Stéphan Clémençon
stephan.clemencon (at)

Postdoctoral position in Brain Computer Interfaces

Faculty of Social Sciences
Maximum Salary: € 4,374 gross/month
Vacancy number: 24.60.08
Closing date: when filled

Job description
This post-doctoral research position in Artificial Intelligence (AI) has a focus on Brain Computer Interfacing (BCI). Brain Computer Interfacing is a new and rapidly growing multi-disciplinary field at the interface between neuro-science and computer-science. We are seeking a highly-motivated technically able candidate to develop novel signal-analysis techniques, mental tasks and subject-training regimes which will push EEG based BCIs to the next level of performance. Current research projects moving towards this goal include; noise-tagging for evoked response BCIs, imagined music/rhythm BCIs, learning spatial/spectral filters, deconvoling overlapping responses, different modalities (tactile/auditory/visual) in BCI, imagined-tapping.

The successful application will be part of the newly formed Donders Institute for Brain, Cognition and Behaviour at the Radboud University of Nijmegen. This position, funded by the large multi-institution and interdisciplinary SmartMix project, BrainGain (see for details), will study new signal analysis techniques and BCI tasks. Facilities and tools to support these studies include real-time motion tracking systems, eye trackers, visual displays, and a vestibular platform. Two state-of the-art MRI systems, whole-head MEG, and an EEG lab are also available.

Preference will be given to candidates with a PhD in Cognitive (Neuro) Science, or signal-processing/machine-learning or related fields, who have a strong motivation to do challenging interdisciplinary research. Further the highly inter-disciplinary nature of BCI research requires an applicant with experience of, or a willingness to learn about, a number of different research areas. Specifically the applicant should have experience in some of: EEG, time-domain and frequency-domain EEG analysis techniques, on-line EEG analysis, evoked and induced response BCIs, neural-correlates of mental tasks, signal-processing techniques, machine-learning, time-series analysis. Strong programming skills, and a demonstrable ability with MatLab is highly desirable.

The applicant will also be expected to help in the design and teaching of BCI courses in the department and supervision of interns (BSc and MSc theses), and PhD students. Further the applicant should have demonstrable experience in conducting scientific research and writing, good collaboration and communication skills. Finally, the application should be able to work in a complex organisation with many partners (Braingain). Pragmatic and productive attitude.

The CAI offers excellent facilities for PhD students and postdocs. It has its own computer support group, electrotechnical and mechanical technicians, secretarial support, and a renowned scientific staff. The CAI is part of the Donders Institute for Brain, Cognition and Behaviour, which is an outstanding research facility on Cognitive Neuroscience. Nijmegen is the oldest city of the Netherlands, with an interesting history dating back to the Roman Empire, nice surrounding scenery (rivers, hills, woods) and a rich cultural life.

Conditions of employment
Maximum employment: 1,0 fte
Maximum salary per month, based on a fulltime employment: € 4,374 gross/month
Salary scale: 11
Duaration of the contract: 3 years.

Other Information
Interested applicants should send CV, statement of background and interests, and names of 2 referees. Review of applications will begin December 31, 2008, and will continue until the position is filled.

Additional Information
dr. Jason Farquhar
Telephone: (+31)024-3611938
E-mail: j.farquhar (at)

prof. dr. P. Desain
Telephone: (+31)024-3615885
E-mail: p.desain (at)

You can apply for the job (mention the vacancy number 24.60.08) before 27 December 2008 by sending your application -preferably by email- to:

Faculty of Social Sciences/P&O
P.O. Box 9104
E-mail: vacancies (at)

3rd Russian Summer School in Information Retrieval (RuSSIR 2009) – Call for Course Proposals

3rd Russian Summer School in Information Retrieval (RuSSIR 2009)
Friday September 11 – Wednesday September 16, 2009
Petrozavodsk, Russia


The 3rd Russian Summer School in Information Retrieval will be held September 11-16, 2009 in Petrozavodsk, Russia. The school is co-organized by the Russian Information Retrieval Evaluation Seminar (ROMIP,, Petrozavodsk State University (, and Karelian Research Centre of the Russian Academy of Sciences ( The first and second RuSSIRs took place in Ekaterinburg in 2007 and Taganrog in 2008, respectively (see and Both events were very successful.

Petrozavodsk, the capital of the Republic of Karelia, was founded in 1703. It is a large industrial and cultural center of the Russian North-West. Petrozavodsk is situated on the shores of Onega Lake, one of the biggest inner lakes in Europe. Karelia is often called “stony lake-forest land” and “the lungs of Europe”, highlighting beautiful landscapes created by countless lakes and rivers and the forest covered land. Petrozavodsk is 400 km away from Saint-Petersburg, an overnight train journey from Saint-Petersburg takes about eight hours. Petrozavodsk State University was founded in 1940 and belongs to the largest educational institutions in the European North of Russia. The university comprises 82 chairs and employs 3,600 faculty/staff members. The total enrollment is more than 19,000 students. IT education and research are one of the main specializations at the university. The Regional Center for New Information Technologies (RCNIT) of PetrSU was the cradle of computer technologies in Karelia and celebrates its 50th anniversary in 2011. PetrSU teams have made remarkable achievements in international student programming

The target audience of the Summer School is advanced graduate and PhD students, post-doctoral researchers, academic and industrial researchers, and developers. The mission of the school is to teach students about a wide range of modern problems and methods in Information Retrieval; to stimulate scientific research in the field of Information Retrieval; and to create an opportunity for
informal contacts among scientists, students and industry professionals. The Russian Conference for Young Scientists in Information Retrieval will be co-organized with the school. RuSSIR2009 will offer 4 or 5 one-week courses and host approximately 100 participants. The working languages of the school are English (preferable) and Russian.

RuSSIR 2009 is co-located with the yearly ROMIP meeting ( and Russian Conference on Digital Libraries 2009 (

The RuSSIR2009 Organizing Committee invites proposals for courses on a wide range of IR-related topics, including but not limited to:
– IR theory and models
– IR architectures
– algorithms and data structures for IR
– text IR
– multimedia (incl. music, speech, image, video, etc.) IR
– natural language techniques in IR tasks
– user interfaces for IR
– Web IR (including duplicate detection, hyperlink analysis, query log
– text mining, information and fact extraction
– mobile applications for IR
– dynamic media IR (blogs, news, WIKIs)
– social IR (collaborative filtering, tagging, recommendation systems)
– IR evaluation.

Each course should consist of five 90-minute-long sessions (normally in five subsequent days). The course may include both lectures and practical exercises in computer labs. A course proposal must contain a brief description of the course (up to 200 words), preferred schedule, prerequisites, equipment needs, a short description of teaching/research experience and contact information of
the lecturer.

RuSSIR2009 organizers will cover travel expenses and accommodation at the school. Lecturers are not paid for their contribution. Details of reimbursement will be negotiated with each lecturer individually. The RuSSIR organizers would highly appreciate if, whenever this is possible, lecturers could find alternative funding to cover travel and accommodation expenses and indicate this possibility in the proposal.

All proposals will be evaluated by the RuSSIR2009 program committee according to the school goals, presentation clarity, lecturer’s qualifications and experience. Topics not featured at previous RuSSIRs are preferred.

Anyone interested in lecturing at RuSSIR2009 is encouraged to submit proposal by email to Pavel Braslavski (pb (at), by January 31, 2009. All submitters will be notified by February 20, 2009 about selection results.

Open positions in Machine Learning, Lille (France)

We would like to advertise that tenure positions for researchers will be opened soon by the French National Research Institute for Computer Science and Control (INRIA,

In Lille, two research groups have stong interest in machine learning
* Sequel (, reinforcement learning
* Mostrare (, structured prediction
4 positions of junior researchers and 1 position of experienced researcher will be open in Lille.
We would also like to mention that opportunities exist for:
* tenure positions for senior researchers in order to create a new research group
* five years positions for senior researchers
* postdoctoral positions
* PhD grants

A thorough description of these opportunities is given on our Web sites.

If you have any question, please get in touch with us, either remi.gilleron (at) (Mostrare), or remi.munos (at), philippe.preux (at) for SequeL.
If you want to apply, it is crucial that you get in touch with us, as early as possible.

Workshop on Advances in Machine Learning for Computational Finance: Call for Contributions

Workshop on Advances in Machine Learning for Computational Finance

Sponsored by the PASCAL 2 Network of Excellence

CALL FOR CONTRIBUTIONS: We solicit submissions for the Advances in Machine Learning for Computational Finance workshop to be held on July 20-21, 2009 at University College London Bloomsbury Campus, London, U.K. Computational finance is a cross-disciplinary field which relies on mathematical finance, numerical methods and computer simulation to make trading, hedging and investment decision, as well as facilitating the risk management of these decisions. Machine learning is concerned with the design and development of algorithm and techniques that extract rules and patterns out of data automatically, by computational and statistical methods.

This workshop brings together researches from machine learning, computational finance, academic finance and the financial industry to discuss problems in finance where machine learning may solve challenging problems and provide an edge over existing approaches. The aim of the workshop is to promote discussion on recent progress and challenges as well as on methodological issues and applied research problems. The emphasis will be on practical problem solving involving novel algorithmic approaches.

Topics of the workshop include (but not limited to):

. *Optimisation methods
. *Reinforcement learning
. *Supervised and semi-supervised learning
. *Kernel methods
. *Bayesian estimation
. *Wavelets
. *Evolutionary computing
. *Recurrent and state space models
. *SVM’s
. *Neural networks
. *Boosting
. *Multi-agent simulation
. *….
. *High frequency data
. *Trading strategies and hedging techniques
. *Execution models
. *Forecasting
. *Volatility
. *Extreme events
. *Credit risk
. *Portfolio management and optimisation
. *Option pricing
. *…

The workshop is a core event of the PASCAL 2 EU Network of Excellence.

SUBMISSION PROCEDURE: We invite the submission of high quality extended abstracts (2 to 4 pages) in the NIPS style ( Abstracts should be sent (in .pdf/.ps/.doc) to the organisers ( , A selection of the submitted abstracts will be accepted as either an oral presentation or poster presentation.


23 Feb 09 – Submission deadline for extended abstracts
30 Mar 09 – Notification of acceptance
20-21 Jul 09 – Workshop at UCL, London, U.K.

David Cliff
University of Bristol

Vince Darley

Vasant Dhar
New York University, Stern School of Business

László Györfi
Budapest University of Technology and Economics

Michael Kearns
University of Pennsylvania

David Leinweber
University of Berkeley, Haas School of Business


David R. Hardoon – University College London
John Shawe-Taylor – University College London
Philip Treleaven -University College London
Laleh Zangeneh – University College London


Nicolò Cesa-Bianchi – Università degli Studi di Milano
Ran El-Yaniv – Technion – Israel Institute of Technology
Samet Gogus – Barclaycard
Yuri Kalnishkan – Royal Holloway, University of London
Jasvindor Kandola – Merrill Lynch
Donald Lawrence – University College London
Giuseppe Nuti – Deutsche Bank
Sandor Szedmak – University of Southampton
Chris Watkins – Royal Holloway, University of London