Job opening at Paris-Sud Orsay

The Department of Computer Science at the University of Paris-Sud has an opening at the assistant professor (maître de conferences) level in the area of machine learning, with a starting date of October 2013. We seek excellent candidates with skills in statistical machine learning, with an interest in applications related to complex systems, e-science, robotics, and/or large-scale ML. The successful candidate will join the Machine Learning and Optimization (TAO) group.

TAO (http://tao.lri.fr), headed by Michèle Sebag and Marc Schoenauer, is a CNRS/INRIA research group at at the University of Paris-Sud, on the beautiful campus of Orsay. The position involves teaching computer science courses at both undergraduate and graduate levels, so a working knowledge of the French language is mandatory. The deadline for the full application is in December, but the hiring procedure includes an administrative phase called “qualification”. Candidates potentially interested in applying should fill out the electronic application before October 25.

More information: posteml@lri.fr

CALL FOR PAPERS: SIMBAD 2013

2nd International Workshop on Similarity-Based Pattern Analysis and Recognition

July 3-5, 2013
York, UK

http://www.dais.unive.it/~simbad

>>> Submission deadline: February 1, 2013 <<< MOTIVATIONS AND OBJECTIVES Traditional pattern recognition techniques are intimately linked to the notion of "feature space." Adopting this view, each object is described in terms of a vector of numerical attributes and is therefore mapped to a point in a Euclidean (geometric) vector space so that the distances between the points reflect the observed (dis)similarities between the respective objects. This kind of representation is attractive because geometric spaces offer powerful analytical as well as computational tools that are simply not available in other representations. However, the geometric approach suffers from a major intrinsic limitation, which concerns the representational power of vectorial, feature-based descriptions. In fact, there are numerous application domains where either it is not possible to find satisfactory features or they are inefficient for learning purposes. In the last few years, interest around purely similarity-based techniques has grown considerably. For example, within the supervised learning paradigm the well-established kernel-based methods shift the focus from the choice of an appropriate set of features to the choice of a suitable kernel, which is related to object similarities. However, this shift of focus is only partial, as the classical interpretation of the notion of a kernel is that it provides an implicit transformation of the feature space rather than a purely similarity-based representation. Similarly, in the unsupervised domain, there has been an increasing interest around pairwise or even multiway algorithms, such as spectral and graph-theoretic clustering methods, which avoid the use of features altogether. By departing from vector-space representations one is confronted with the challenging problem of dealing with (dis)similarities that do not necessarily possess the Euclidean behavior or not even obey the requirements of a metric. The lack of such properties undermines the very foundations of traditional pattern recognition theories and algorithms, and poses totally new theoretical/computational questions and challenges. The aim of this workshop, which follows the one held in Venice in 2011 (http://www.dais.unive.it/~simbad/2011/), is to consolidate research efforts in this area, and to provide an informal discussion forum for researchers and practitioners interested in this important yet diverse subject. We aim at covering a wide range of problems and perspectives, from supervised to unsupervised learning, from generative to discriminative models, and from theoretical issues to real-world applications. Original, unpublished papers dealing with these issues are solicited. Topics of interest include (but are not limited to): - Embedding and embeddability - Graph spectra and spectral geometry - Indefinite and structural kernels - Game-theoretic models of pattern recognition - Characterization of non-(geo)metric behavior - Foundational issues - Measures of (geo)metric violations - Learning and combining similarities - Multiple-instance learning - Applications PAPER SUBMISSION All papers (not exceeding 16 pages) must be submitted electronically at the conference website (http://www.dais.unive.it/~simbad/2013/). All submissions will be subject to a rigorous peer-review process. Accepted papers will appear in the workshop proceedings, which will be published in Springer's Lecture Notes in Computer Science (LNCS) series. In addition to regular, original contributions, we also solicit papers (in any LaTeX format, no page restriction) that have been recently published elsewhere. These papers will undergo the same review process as regular ones: if accepted, they will be presented at the workshop but will not be published in the workshop proceedings. Submission implies the willingness of at least one of the authors to register and present the paper, if accepted. IMPORTANT DATES Paper submission: February 1, 2013 Notifications: March 15, 2013 Camera-ready due: April 25, 2013 Workshop: July 3-5, 2013 ORGANIZATION Program Chairs Edwin Hancock, University of York, UK Marcello Pelillo, University of Venice, Italy Steering Committee Joachim Buhmann, ETH Zurich, Switzerland Robert Duin, Delft University of Technology, The Netherlands Mario Figueiredo, Technical University of Lisbon, Portugal Edwin Hancock, University of York, UK Vittorio Murino, Italian Institute of Technology, Italy Marcello Pelillo (chair), University of Venice, Italy Program Committee Maria-Florina Balcan, Georgia Institute of Technology, USA Joachim Buhmann, ETH Zurich, Switzerland Terry Caelli, NICTA, Australia Tiberio Caetano, NICTA, Australia Umberto Castellani, University of Verona, Italy Luca Cazzanti, University of Washington, Seattle, USA Robert Duin, Delft University of Technology, The Netherlands Aykut Erdem, Hacettepe University, Ankara, Turkey Francisco Escolano, University of Alicante, Spain Mario Figueiredo, Technical University of Lisbon, Portugal Ana Fred, Technical University of Lisbon, Portugal Mehmet Gonen, Aalto University School of Science, Finland Marco Gori, University of Siena, Italy Edwin Hancock, University of York, UK Robert Krauthgamer, Weizmann Institute of Science, Israel Marco Loog, Delft University of Technology, The Netherlands Marina Meila, University of Washington, Seattle, USA Vittorio Murino, Italian Institute of Technology, Italy Marcello Pelillo, University of Venice, Italy Massimiliano Pontil, University College London, UK Antonio Robles-Kelly, NICTA, Australia Fabio Roli, University of Cagliari, Italy Samuel Rota Bulo', University of Venice, Italy Volker Roth, University of Basel, Switzerland John Shawe-Taylor, University College London, UK Andrea Torsello, University of Venice, Italy Richard Wilson, University of York, UK Lior Wolf, Tel Aviv University, Israel

FACULTY MEMBER IN MACHINE LEARNING OR STATISTICS – UCL, UK

The Gatsby Computational Neuroscience Unit at University College London is looking to recruit a junior or senior level faculty member in machine learning or statistics. We especially seek candidates who work in probabilistic or statistical machine learning.

The Gatsby Unit was set up at UCL in 1998 as a research institute devoted to machine learning and theoretical neuroscience. We have core funding for five faculty and for associated postdocs and PhD students. PIs can raise additional funds through grants. We have no undergraduate programme, so only teaching and supervision of graduate-level Gatsby students is required.

Along with the existing machine learning at Gatsby, led by Arthur Gretton and Maneesh Sahani, UCL offers a rich environment across the breadth of the field.

Activities in these areas are anchored by the Centre for Computational Statistics and Machine Learning which is directed by Mark Girolami, involving the departments of Computer Science (including David Barber, Gabriel Brostow, Mark Herbster, Massimiliano Pontil, Sebastian Riedel, David Silver and John Shawe-Taylor), Statistics (including Alex Beskos, Sofia Olhede and Ricardo Silva) and the Gatsby Unit itself. The Unit also has close ties with groups in Engineering at Cambridge (Zoubin Ghahramani, Carl Rasmussen, who were both formerly Unit members), Statistics in Oxford (Yee Whye Teh, who has just left) and beyond.

The Unit is currently located in a leafy haven in Queen Square, London. We will move in 2014 to a new building to be shared with the Sainsbury Wellcome Centre for Neural Circuits and Behaviour, a development which is expected to offer many opportunities for interaction.

The Unit offers internationally competitive salaries. These range from a starting point of £41,639 for a lecturer to a top point of £52,706 for a Senior Lecturer/Reader. Professorial posts have a starting point of £59,304 per annum but are negotiable.

In addition a London Allowance of £2,806 per annum is payable and a market supplement is available for exceptional candidates in line with international ranges.

Applications must be made online via the UCL job vacancies website: http:// www.ucl.ac.uk/hr/jobs/. Please be sure to attach to your online application a copy of your CV, statement of research interests and accomplishments, a teaching statement, and full contact details (including e-mail addresses) for three academic referees.

For further information, please see www.gatsby.ucl.ac.uk/vacancies/FacultyJD.pdf

For informal enquiries, please contact Peter Dayan at dayan@gatsby.ucl.ac.uk.

The closing date for applications is 20 December 2012

Interviews will be held in the New Year

PhD in Healthcare analytics & PhD in Biometrics – Surrey University

PhD in Healthcare analytics
Electronic health records contain a wealth of information that has not been fully exploited. Today, it is possible to retrieve millions of patient records over time, across vendors, and clinical practices. This research project aims to develop a set of statistical tools for processing population-scale health records, that is, in the order of millions of patient records. Although a number of statistical software packages exist, e.g., there is still room for improvement, e.g., designing better algorithms that are capable of handling sampling bias, structural noise, or under-sampled data; handling covariates or confounding factors; exploiting temporal logics; and, efficiently retrieving patient records.The work will concentrate upon novel pattern recognition, machine learning, and data-mining techniques. The student is expected to be able to use or modify existing statistical tools or methodologies in order to solve novel problems posed by healthcare informatics.
To find out more: http://personal.ee.surrey.ac.uk/Personal/Norman.Poh/vacancy.php

PhD in Biometrics
The project addresses the problem of evaluation and testing biometrics systems. The current practice is to evaluate the performance of biometrics systems on standard data sets, using simple criteria, such as misclassification rate, or false positive and false rejections rates. This does not gauge the sensitivity of the systems to various forms of degradation, such as registration errors, illumination problems, noise, focus and motion blur, etc. The aim of the project will be to develop novel performance metrics which will capture the effect of degradations on the system performance. The challenge will be to build models of degradation processes, and to develop techniques for generating realistic synthetic data that could be used for biometric system evaluation. The possibility of calibrating the performance measures by means of synthetic to real data association models will be investigated.
To find out more: http://personal.ee.surrey.ac.uk/Personal/Norman.Poh/vacancy.php

Postdoctoral researcher – Katholieke Universiteit Leuven (KU Leuven), Belgium

The Language Intelligence & Information Retrieval (LIIR) group (http://www.cs.kuleuven.be/groups/liir/, Department of Computer Science, Human Computer Interaction section) at the Katholieke Universiteit Leuven (KU Leuven), Belgium has an open position for a

Postdoctoral researcher
(Full time, 1 year with possibility for extension)
In the field of

multimedia retrieval, machine learning and retrieval models.

The position is financed by the European TOSCA-MP project (Task-Oriented Search and Content Annotation for Media Production – EU FP7-287532 – http://tosca-mp.eu/) in which LIIR collaborates with important European teams specialized in multimedia retrieval and with European broadcasters.

The successful applicant has:

• Completed the PhD with success as evidenced by multiple publications in proceedings of important conferences and journals in the field of information retrieval;
• A master degree in computer science, electrical engineering, mathematics, physics or a related discipline;
• Acquired a profound knowledge and practical experience in machine learning, data mining and information retrieval; experience with aggregated search, reranking algorithms, optimization methods, structured input and output spaces, graphical models, probabilistic inference, message-passing algorithms, and related techniques is an advantage;
• Good programming skills (e.g., Java, C++, MATLAB, Python);
• Excellent English language skills (written and spoken);
• Good communication skills especially for guiding master and PhD students;
• The capability to work independently and in a team;
• A profound interest in multimedia information retrieval.

Please submit your application and cv in electronic form to Prof. Marie-Francine Moens (sien.moens@cs.kuleuven.be). The starting date is January 1, 2013. We are accepting applications until November 15, 2012, or until the position fills.

Junior researcher – Katholieke Universiteit Leuven (KU Leuven), Belgium

The Language Intelligence & Information Retrieval (LIIR) group (http://www.cs.kuleuven.be/groups/liir/, Department of Computer Science, Human Computer Interaction section) at the Katholieke Universiteit Leuven (KU Leuven), Belgium has an open position for a

Junior researcher
(Full time, 4 years)
In the field of

machine reading, natural language processing and machine learning,
leading to a PhD in Engineering, Computer Science.

The PhD fellowship is sponsored by the project SAEL-RMR (A Synergetic Approach to Extraction, Learning and Reasoning for Machine Reading – G.0356.12), in which LIIR collaborates with the artificial intelligence group of KU Leuven and the Web intelligence group of Ghent University (Belgium), and by the European project (Machine Understanding for Interactive StorytElling – EU FP7-296703 – http://www.muse-project.eu/), in which LIIR collaborates with famous European groups with regard to knowledge management, storytelling, cognitive science and processing of medical texts.
The goal of the PhD research is to perform research in an area known as machine reading, i.e., automatically extracting knowledge from text. More specifically, the focus is on algorithms that use natural language processing techniques for reading the textual information and to design inferencing techniques that reason over uncertain evidences in order to improve the accuracy of the knowledge acquisition. The developed technologies will be evaluated on biomedical texts.

The successful applicant has:

• Completed the master degree with success (cum laude) as evidenced by excellent grades and possible publications;
• A master degree in computer science, electrical engineering, mathematics, physics or a related discipline;
• Acquired knowledge and practical experience in machine learning, data mining and information retrieval; knowledge of optimization methods, language modeling, structured input and output spaces, graphical models, probabilistic inference, message-passing algorithms and related techniques is an advantage;
• Good programming skills (e.g., Java, C++, MATLAB, Python);
• Excellent English language skills (written and spoken);
• Good communication skills;
• The capability to work independently and in a team;
• A profound interest in machine understanding of text.

Please submit your application, cv and grade transcripts in electronic form to Prof. Marie-Francine Moens (sien.moens@cs.kuleuven.be). The starting date is of the PhD position January 1, 2013. We are accepting applications until November 15, 2012, or until the position fills.

Two Post-Doctoral positions in Machine Learning and Computational Biology

AVAILABLE AT LRI, CNRS/INRIA-Saclay, University of Paris Sud and IBISC, University of Evry & GENOPOLE, FRANCE

Postdoc 1:

Within the context of a collaboration between the LRI laboratory from University of Paris Sud in the new Saclay campus, the IBISC laboratory from University of Evry (Genopole) and a group of biologists from Paris Descartes University, we seek a postdoctoral candidate in machine learning applied to bioinformatics for one year with a possible extension to two years. He/she will investigate learning methods for inferring the human protein-protein interaction network related to protein CFTR (cystic fibrosis). A sound knowledge in machine learning and a motivation for bioinformatics are required.

Postdoc 2:

Within a collaboration with the French Institute for Radioprotection and Nuclear Safety (IRSN), the IBISC laboratory is seeking postdoctoral candidates for a two-year position with a strong background in statistical learning and systems biology. The successful applicant will participate to the IRSN research project ROSIRIS devoted to the radiotherapy side effects and particularly to the understanding of endothelium dysfunctions in normal tissues following ionizing radiation exposure. He/she will investigate new learning algorithms for dimension reduction and macromolecular network inference from kinetics of primary endothelial cells protein abundances. A sound knowledge in dynamical graphical models and/or kernel approaches will be appreciated.

Application

The successful candidates will have a Ph.D. in Computer Science, Mathematics or Bioinformatics with a strong publication record. Applications for these roles should include a statement of research experience and interests, a CV, two reference letters from appropriate academic sources and the two best publications.

Each position will preferably start before the end of 2012 but the beginning of the postdoc position can be discussed.

Please send your applications in electronic form to :

Postdoc 1:
Prof. Florence d’Alché-Buc: florence.dalche@ibisc.fr, Prof. Christine Froidevaux: christine.froidevaux@lrir,

Postdoc 2:
Prof. Florence d’Alché-Buc: florence.dalche@ibisc.fr, Dr Olivier Guipaud (Olivier.guipaud@irsn.fr) and Dr Farida Zehraoui (farida.zehraoui@ibisc.fr).

Website of the group: http://amis-group.fr/
Website of LRI: http://www.lri.fr

CVMP2012 – 5th & 6th December, Vue Leicester Square, London

Early bird REGISTRATION NOW OPEN – Get £100 off the full ticket price.

There are a limited number of Early bird tickets so make sure you get yours soon. Buy from the ‘Register’ page on the website. (Ticket price includes access to all areas & includes refreshments, lunch & entry to the Drinks Reception on Wednesday 5th Dec)
Short paper deadline – 1 week to go

Enter short papers by Friday 19th October.

www.cvmp-conference.org

CVMP 2012 – Short Paper submissions ending soon…

Short Paper deadline fast approaching!

The submission deadline for full papers is 19th October 2012, so you only have a week left.
Accepted papers will appear in the conference programme. For more information about online submission, please go to the CVMP website. (http://s2814.t.en25.com/e/er?utm_campaign=EMEA%20EVENT%2012%20Oct%20CVMP%20Short%20Paper%20Reminder&utm_medium=email&utm_source=Eloqua&s=2814&lid=96&elq=2d2deba9aa9b4669b739cd2a65c27763)

Submission Deadline: 19th October 2012

Earlybird Registration

Registration for CVMP is now open! Earlybird tickets are on sale now, but hurry! There are only a
limited number of Earlybird tickets available – Register for CVMP 2012. (http://www.cvmp-conference.org/Register?utm_campaign=EMEA%20EVENT%2012%20Oct%20CVMP%20Short%20Paper%20Reminder&utm_medium=email&utm_source=Eloqua&elq=2d2deba9aa9b4669b739cd2a65c27763&elqCampaignId=93)
We look forward to hearing from you.

Jan Kautz, UCL
Programme Chair

Reminder: Multi-Trade-offs in Machine Learning, NIPS-2012 workshop

CALL FOR ABSTRACTS AND OPEN PROBLEMS
Multi-Trade-offs in Machine Learning
NIPS-2012 Workshop, Lake Tahoe, Nevada, US
https://sites.google.com/site/multitradeoffs2012/
December 7 or 8, 2012

———————————————-
We invite submission of abstracts and open problems to Multi-Trade-offs in Machine Learning NIPS-2012 workshop.
IMPORTANT DATES
Submission Deadline: October 16.
Travel Support Application Deadline: October 23.
Notification of Acceptance: October 30.
More details are provided below.

———————————————–
Abstract

One of the main practical goals of machine learning is to identify relevant trade-offs in different problems, formalize, and solve them. We have already achieved fairly good progress in addressing individual trade-offs, such as model order selection or exploration-exploitation. In this workshop we would like to focus on problems that involve more than one trade-off simultaneously. We are interested both in practical problems where “multi-trade-offs” arise and in theoretical approaches to their solution. Obviously, many problems in life cannot be reduced to a single trade-off and it is highly important to improve our ability to address multiple trade-offs simultaneously. Below we provide several examples of situations, where multiple trade-offs arise simultaneously. The goal of the examples is to provide a starting point for a discussion, but they are not limiting the scope and any other multi-trade-off problem is welcome to be discussed at the workshop.

Multi-trade-offs arise naturally in interaction between multiple learning systems or when a learning system faces multiple tasks simultaneously; especially when the systems or tasks share common resources, such as CPU time, memory, sensors, robot body, and so on. For a concrete example, imagine a robot riding a bicycle and balancing a pole. Each task individually (cycling and pole balancing) can be modeled as a separate optimization problem, but their solutions has to be coordinated, since they share robot resources and robot body. More generally, each learning system or system component has its own internal trade-offs, which have to be balanced against the trade-offs of other systems, whereas shared resources introduce external trade-offs that enforce cooperation. The complexity of interaction can vary from independent systems sharing common resources to systems with various degrees of relation between their inputs and tasks. In multi-agent systems communication between the agents introduces an additional trade-off.
We are also interested in multi-trade-offs that arise within individual systems. For example, model order selection and computational complexity [1], or model order selection and exploration-exploitation [2]. For a specific example of this type of problems, imagine a system for real-time prediction of the location of a ball in table tennis. This system has to balance between at least three objectives that interact in a non-trivial manner: (1) complexity of the model of flight trajectory, (2) statistical reliability of the model, (3) computational requirements. Complex models can potentially provide better predictions, but can also lead to overfitting (trade-off between (1) and (2)) and are computationally more demanding. At the same time, there is also a trade-off between having fast crude predictions or slower, but more precise estimations (trade-off between (3) and (1)+(2)). Despite the complex nature of multi-trade-offs, there is still hope that they can be formulated as convex problems, at least in some situations [3].

References:
[1] Shai Shalev-Shwartz and Nathan Srebro. “SVM Optimization: Inverse Dependence on Training Set Size”, ICML, 2008.
[2] Yevgeny Seldin, Peter Auer, François Laviolette, John Shawe-Taylor, and Ronald Ortner. “PAC-Bayesian Analysis of Contextual Bandits”, NIPS, 2011.
[3] Andreas Argyriou, Theodoros Evgeniou and Massimiliano Pontil. Convex multi-task feature learning. Machine Learning, 2008, Volume 73, Number 3.

Call for Contributions

We invite submission of abstracts and open problems to the workshop. Abstracts and open problems should be at most 4 pages long in the NIPS format (appendices are allowed, but the organizers reserve the right to evaluate the submissions based on the first 4 pages only). Selected abstracts and open problems will be presented as talks or posters during the workshop. Submissions should be sent by email to seldin@tuebingen.mpg.de.

IMPORTANT DATES

Submission Deadline: October 16.

Notification of Acceptance: October 30.

EVALUATION CRITERIA
• Theory and application-oriented contributions are equally welcome.
• All the submissions should indicate clearly at least two non-trivial trade-offs they are addressing.
• Submission of previously published work or work under review is allowed, in particular NIPS-2012 submissions. However, for oral presentations preference will be given to novel work or work that was not yet presented elsewhere (for example, recent journal publications or NIPS posters). All double submissions must be clearly declared as such!

Invited Speakers

Shai Shalev-Shwartz, The Hebrew University of Jerusalem
Jan Peters, Technische Universitaet Darmstadt and Max Planck Institute for Intelligent Systems
Csaba Szepesvari, University of Alberta

Organizers

Yevgeny Seldin, Max Planck Institute for Intelligent Systems and University College London
Guy Lever, University College London
John Shawe-Taylor, University College London
Koby Crammer, The Technion
Nicolò Cesa-Bianchi, Università degli Studi di Milano
François Laviolette, Université Laval (Québec)
Gábor Lugosi, Pompeu Fabra University
Peter Bartlett, UC Berkeley and Queensland University of Technology

Sponsors

We are grateful for receiving support from the PASCAL network.

If you would also like to sponsor this event, please, contact seldin@tuebingen.mpg.de.

Tentative Schedule
7:30 – 7:35 Opening remarks
7:35 – 8:20 Invited Talk
8:20 – 8:50 Two Contributed Talks
8:50 – 9:10 Break
9:10 – 9:55 Invited Talk
9:55 – 10:30 Open Problems Session
10:30 – 15:30 Break
15:30 – 16:15 Invited Talk
16:15 – 16:25 Break
16:25 – 17:00 Two Contributed Talks
17:00 – 18:30 Posters
18:30 – 19:00 Workshop Summary and Open Discussion