CFP: META-NET Workshop on “Context in MT”, 14 June 2011

Call for abstracts and participation:

CONTEXT IN MACHINE TRANSLATION
META-NET workshop in ICANN 2011 conference
Espoo, Finland, 14th of June, 2011
Abstract submission: 16th of March, 2011

=== BACKGROUND: ICANN CONFERENCE ===

ICANN 2011 conference brings together researchers from computational
modeling, including machine learning and neural networks as well as
different scientific disciplines in which these methods are applied.
In the ICANN 2011 conference, there will be six keynotes by
renowned scientists:

– Tom Griffiths, University of California Berkeley,
– Riitta Hari, Aalto University,
– Geoffrey Hinton, University of Toronto,
– Aapo Hyvärinen, University of Helsinki,
– John Shawe-Taylor, University College London, and
– Josh Tenenbaum, Massachusetts Institute of Technology.

=== MACHINE TRANSLATION RESEARCH AND CONTEXT ===

Machine Translation can be considered to be one of the most
challenging tasks computer science has ever taken. Statistical methods
have been increasingly successful in providing efficiently MT
solutions for many language pairs. However, there is a lot of room for
improvement regarding the quality of translations. Prototypical
sentences are translated well but in certain situations the end result
is far from expected. One central reason for the failures is that
current systems take the context into account only in a limited
manner.

In natural language processing, the context of use has a considerable
impact on the understanding process. It can refer to multiple kinds of
meta-data, including information on the document type, domain, genre
and medium used. Automatic machine translation systems typically
restrict the considered context to one sentence or smaller parts of
it.

=== WORKSHOP INFORMATION ===

META-NET Network of Excellence organizes a workshop on Context in
Machine Translation to foster exchange of ideas and results in this
area. The notion of context is meant to be understood broadly,
including other modalities (like vision) in addition to the textual
contexts. Therefore, we expect participants, e.g., from machine
translation, machine learning, vision research and cognitive systems
communities.

The workshop takes place on Tuesday, 14th of June, 2011 at Aalto
University School of Science as a part of ICANN 2011 conference
programme (see http://www.cis.hut.fi/icann11/ for details).

In the workshop, the task and data for Context in Machine Translation Challenge will be
introduced.

=== CALL FOR CONTRIBUTIONS ===

We invite presentations related to the domain of the workshop. This is
an interdisciplinary event and therefore contributions from different
relevant disciplines including language technology, machine learning and cognitive
systems are welcome.

The registration information to ICANN 2011 including conference fees
is available at http://www.cis.hut.fi/icann11/.

If you would like to give a presentation, please send the title and and (extended)
abstract of your presentation in pdf format by 16th of March, 2011 to timo.honkela@tkk.fi
and jaakko.j.vayrynen@tkk.fi. The book of abstracts will be available in the workshop.
The workshop programme will be announced by the end of March, 2011.

If you wish to participate to the workshop without a presentation, please register
normally to the conference and send an e-mail to
jaakko.j.vayrynen(at)tkk.fi by 10th of June, 2011. Please note
15th of April as the deadline for early bird registration.

Research position to work on autonomously motivated exploration and skill acquisition in reinforcement learning

For the EU-funded project CompLACS (Composing Learning for Artificial Cognitive Systems) we are looking for a highly motivated post doctoral researcher in machine learning/reinforcement learning to develop well founded (mathematical) models for autonomous exploration and skill acquisition.

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

This position will be filled in April 2011 for up to four years. Interviews will be conducted in mid March. Depending on your qualification, salary is in the range 35000-45000 EUR per year (after all social and insurance benefits and taxes this is net 24000-30000 EUR). Highly qualified PhD candidates might 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

ECML PKDD 2011 – Call For Discovery Challenge Data/Tasks

Call For Discovery Challenge Data/Tasks

ECML-PKDD 2011: European Conference on Machine Learning and Principles and
Practice of Knowledge Discovery in Databases

September 5-9, 2011, Athens, Greece

http://www.ecmlpkdd2011.org/

The organizing committee of ECML-PKDD 2011 invites the submission of
proposals for the tasks and datasets to be used in ECML-PKDD Discovery
Challenge 2011. The Discovery Challenge provides a venue for the
collaborative exploration and evaluation of novel and interesting
tasks in the areas of Machine Learning and Knowledge Discovery.

The outcome of the Discovery Challenge will be presented in a dedicated
workshop on the first or the last day of the conference, where the
winning and other selected participations will be presented.

* Proposal Guidelines

Submitted proposals should be self-contained and should include at least
the following information:

– Title and abstract of the proposed challenge.
– Contact information of organizer(s).
– General description of the proposed challenge.
– Description of the dataset used in the proposed challenge.
– Description of at least two tasks in which the Discovery Challenge
participants will compete.
– Description of the evaluation procedure and metrics.
– Short biography of organizers, mentioning relevant experience.

The submitted proposals should also consider the following points:

– The dataset should be made publicly available after the end of the
challenge to facilitate future use and repeatability of experiments.
– The dataset should be sufficiently large.
– The tasks should address novel and interesting research problems, which
do not require significant domain-specific knowledge from participants.
– The evaluation procedure should be described clearly.

We encourage organizers to seek awards and prizes for the winning
participants in order to attract participation. The acceptance of a
proposed Discovery Challenge, however, does not depend on the
provision of an award or prize.

* Important Dates

Mar 21, 2011: Discovery Challenge proposal deadline.
Mar 28, 2011: Notification of acceptance.
Apr 11, 2011: Discovery Challenge announcement: Description of tasks, dataset and metrics become available.
Jul 10, 2011: Submission of Discovery Challenge results and reports in
appropriate format.
Jul 21, 2011: Notification of acceptance.
Jul 28, 2011: Submission of Camera Ready Copy for selected submissions.
Sep 5 or 9, 2011: Discovery Challenge workshop at ECML-PKDD 2011.

* Submission of Proposals

Discovery Challenge proposals should be submitted in PDF form by email to
both Discovery Challenge chairs. All submissions will be acknowledged by
email.

* Contact information

If you have any further questions, please do not hesitate to contact us.
We are looking forward to your proposals.

Alexandros Kalousis (Alexandros.Kalousis(at)unige.ch), University of Geneva
Vassilis Plachouras (vplachouras(at)aueb.gr), Athens Univ. of Economics and
Business

MSR PhD studentship available at Edinburgh

A Microsoft Research funded studentship is available at the School of Informatics, University of Edinburgh, to work on machine learning methodology for systems biology. Specifically, the research will involve modelling plants’ circadian clocks; it will be jointly supervised by Dr Guido Sanguinetti (Informatics, Edinburgh) and Prof Chris Bishop (MSR, Cambridge), and will involve collaboration with Prof Andrew Millar’s group at the Centre for Systems Biology at Edinburgh.

For further information and how to apply, please see
http://www.jobs.ac.uk/job/ACG192/microsoft-research-studentship/

RA in Computational Complexity and Behavioural Evolution, Sheffield UK

Research Associate in Computational Complexity and Behavioural Evolution
University of Sheffield, UK
Duration: 18-months with potential for extension
Deadline: 25th March 2011

This post will develop novel approaches to understanding the evolution of animal behaviour, by applying computational complexity and information theoretic approaches to the analysis of mechanisms for implementing animal behaviour. The classic approach to understanding behavioural evolution is in terms of models of optimal behaviour. Optimality theory is invaluable as a benchmark to assess a given behaviour or behavioural model against, but typically requires unrealistic limitations on the behaviour under consideration, and hence results in ‘complex models for simple environments’ [2]. However, real animals inhabit complex environments, and use simple rules of thumb to deal with them [1]. There is clearly a trade-off between the marginal fitness gain from having a more and more complex behavioural model for an environment, and the fitness costs of the additional resources required for implementing that model. To date, the approach to understanding this trade-off has been somewhat heuristic. This project will seek to apply computational complexity theory and information theory in an attempt to gain a more quantitative understanding of the trade- off between behavioural complexity and behavioural optimality.
It is expected that the successful candidate will quickly take a lead in developing a research programme and securing funding to continue it. Researchers who have demonstrated an ability to direct their own research programme, whether during doctoral studies or subsequently, are therefore particularly encouraged to apply. We particularly welcome applications from theoretical computer scientists, theoretical physicists, and mathematicians. A demonstrated interest in biology would be an advantage.
The successful candidate will become part of the newly established Behavioural and Evolutionary Theory Lab at the University of Sheffield, Department of Computer Science, under the directionof Dr James Marshall (http://staffwww.dcs.shef.ac.uk/people/J.Marshall/). It is anticipated that there will be opportunities for interaction with the Modelling Animal Decisions research group directed by Professors Alasdair Houston and John McNamara (Mathematics) at the University of Bristol (http://www.bristol.ac.uk/biology/research/behaviour/mad/).

References
—
[1] Gigerenzer, G., Todd, P.M. et al. (1999) Simple Heuristics that Make Us Smart. Oxford University Press.
[2] McNamara, J.M and Houston, A.I. (2009) Integrating function and mechanism. Trends in Ecology and Evolution 24, 670-675.

For access to the full job advert, visit http://staffwww.dcs.shef.ac.uk/people/J.Marshall/lab/Join_Us.html

2 postdoctoral positions at Cambridge

UNIVERSITY POSTDOCTORAL RESEARCH FELLOWSHIPS IN STATISTICS (2 POSTS)
Department of Pure Mathematics and Mathematical Statistics
Salary: GBP 27,319-?35,646 pa
Limit of Tenure applies*

Applications are invited for two Postdoctoral Research Fellowships in
Statistics, to be held in the Statistical Laboratory.

Informal enquiries can be directed to Professor A.P. Dawid (email:
a.p.dawid(at)statslab.cam.ac.uk).
Further particulars and full details of how to apply can be found at
www.statslab.cam.ac.uk/Vacancies/.

* Limit of tenure: 3 years
Quote Reference: LF07669, Closing Date: 4 March 2011

THREE Postdoctoral Fellowships in Robotics, Machine Learning and Animation @ University of Edinburgh, UK

We have THREE postdoctoral openings in the field of Robotics and Animation, with an emphasis on Machine Learning techniques for adaptive representation, dynamic planning and motion synthesis in high dimensional, anthropomorphic robotic systems and articulated full-physics animation. We are looking for highly motivated individuals with a strong publication record and solid background in the fundamentals of machine learning, planning and control in robots and/or experience with physically based graphics and animation. The three posts are expected to work in a complementary manner towards the goals of the larger, multi site project: Topology based Motion Synthesis, details of which can be found on the application website.

Posts are tenable for a maximum of three years starting April 2011 and attract a salary in the UE07 pay scale: £29,972 – £35,788 commensurate to experience.
Please refer to details of individual posts, requirements and application procedure by going to www.jobs.ed.ac.uk and entering the appropriate REFERENCE number and clicking on ‘Further Information’:

1. Robotics and Machine Learning (enquiry: Prof. Sethu Vijayakumar) — REFERENCE No. 3014034, Closing date: 4 March 2011
2. Autonomous Robotics (enquiry: Dr. Subramanian Ramamoorthy) – – REFERENCE No. 3014035, Closing date: 4 March 2011
3. Robotics and Animation (enquiry: Dr. Taku Komura) — REFERENCE No. 3014081, Closing date: 10 March 2011

Please upload statement of interest, CV, references and other details by following the online application at: www.jobs.ed.ac.uk
Interviews: Mid March 2011

Reader/Professorship in Computational/Systems Biology at University College London

University College London has an opening for a Reader/Professorship in
Computational/Systems Biology.

Please see the following link for more information:

http://www.jobs.ac.uk/job/ACF512/reader-professorship-in-computational-systems-biology/

PASCAL is now on Twitter

If you would like to be kept up to date with latest news from PASCAL, you can follow us on Twitter.
The Twitter feed will contain the latest Job Vacancies, Calls for Participation and latest news from the network. Alternatively you can subscribe to our RSS Feed.

If you have any other items that you would like added onto the website or our Twitter Feed, please email rebecca.martin (at) cs.ucl.ac.uk.

Announcing MLDATA

Dear Pascal researchers,

we are proud to announce mldata, the machine learning data set
repository at http://mldata.org.

mldata is a community website aimed at exchanging data sets. Compared
to existing sites, the emphasis lies on community. That means that
anyone can upload data, comment on existing data sets, contribute
solutions to existing data sets, discuss topics in the forum, and in
general easily interact with other users.

mldata is organized into four main types of objects:

* Data – just raw data
* Task – learning tasks defined on data sets
* Method – a machine learning method, can be applied to a Task
* Challenge – a set of Tasks defining a challenge

In principle, any kind of data can be uploaded, but mldata can parse
some data formats like ARFF, CSV, and that used libsvm and other SVM
solvers. For such data sets, more functionality is available like
automatic conversion to other data sets.

Other features include automatic evaluation of solutions for tasks
using one of a large number of already available performance measures,
but of course we’re glad to add any user contributed performance measure.

So have a look, and let us know what you think on the mldata forum!

Mikio Braun – on behalf of the mldata team.

mldata is sponsored by the Pascal2 Network of Excellence.