The PASCAL Probabilistic Inference Challenge evaluates algorithms for approximate inference in graphical models. Tasks involve calculating the partition function, marginals and MAP assignments.
The deadline has now been extended and the final date is March 15th.
You are all invited to submit your algorithms and participate in the competition.
For more details please see: www.cs.huji.ac.il/project/PASCAL/
CoMPLEX is currently advertising two fellowships to work with the 2020 Science research programme.
2020 Science is a collaborative research programme based at the University College London, University of Oxford, and Microsoft Research, Cambridge that will involve up to 17 fellows over 5 years.
The programme is focused on fostering the creation of a new generation of interdisciplinary research leaders; scientists who are able to apply and develop computational and mathematical modelling approaches to advance our understanding of complex natural systems.
The suitable candidates will be highly motivated researchers with a PhD in a relevant area of science, such as: mathematical or computational biology, computer science or biology. Research experience of mathematical or computational modelling of complex natural systems is essential, as well as the ability to conduct and complete research projects, as witnessed by published peer-reviewed work.
The post duration will be 24 months.
The deadline for applications is noon on Friday 16th March, and full details about the application process for these positions can be found at http://www.ucl.ac.uk/complex/vacancies
Further details about the 2020 Science programme can be found at http://www.2020science.net/.
http://bmvc2012.surrey.ac.uk/
The British Machine Vision Conference (BMVC) is one of the major international conferences on machine vision and related areas. Organized by the British Machine Vision Association, the 23rd BMVC will be held in Guildford UK, at the University of Surrey.
Authors are invited to submit full-length high-quality papers in image processing and machine vision. Papers covering theory and/or application areas of computer vision are invited for submission. Submitted papers will be refereed on their originality, presentation, empirical results, and quality of evaluation.
All papers will be reviewed *doubly blind*, normally by three members of our international programme committee. Please note that BMVC is a single track meeting with oral and poster presentations and will include two keynote presentations and two tutorials.
Topics include, but are not limited to:
• Statistics and machine learning for vision
• Stereo, calibration, geometric modelling and processing
• Person, face and gesture tracking
• Object and activity recognition
• Motion, flow and tracking
• Segmentation and feature extraction
• Model-based vision
• Image processing techniques and methods
• Texture, shape and colour
• Video analysis
• Document processing and recognition
• Vision for quality assurance, medical diagnosis, etc.
• Vision for visualization, interaction, and graphics
The conference will include company exhibits and a demonstration session. All oral presentations will be recorded and hosted on videolectures.net. As with previous years, the best papers of the conference will be invited to submit a journal publication to IJCV.
Conference Chairs: Dr John Collomosse
Dr Krystian Mikolajczyk
Prof Richard Bowden
Invited Speakers: Prof Stan Sclaroff, Boston University, US
Prof Jiri Matas, Czech Technical University, Prague
Tutorials: Large-scale and larger-scale image search, Dr Herve Jegou, INRIA RENNES, France.
MAP inference in Discrete Models, Dr Pushmeet Kohli, Microsoft Research, UK
Important Dates:
26 April 2012 Abstracts due
3 May 2012 Full paper submissions due
6 July 2012 Author notifications
1 August 2012 Camera ready papers due
3-7 September 2012 Conference
See http://bmvc2012.surrey.ac.uk/ for more details
BMVC2012 is sponsored by Pascal2, Microsoft Research and Stemmer Imaging
===========================================
SECOND CALL FOR PAPERS
===========================================
ACL Workshop ExProM 2012
Extra-propositional aspects of meaning in computational linguistics
Organised by the University of Antwerp and Saarland University
Colocated with ACL 2012
Sponsored by PASCAL2
July 2012, Jeju Island, Korea
http://www.clips.ua.ac.be/exprom2012
===========================================
Papers are invited for the one-day workshop to be held in Jeju Island, Korea, on July 13, 2012.
Until recently, research in Natural Language Processing (NLP) has focused predominantly on propositional aspects of meaning. For example, semantic role labeling, question answering or text mining tasks aim at extracting information of the type “who does what, when and where”. However, understanding language involves also processing Extra-Propositional Aspects of Meaning (EPAM), such as factuality, uncertainty, or subjectivity, since the same propositional meaning can be presented in a diversity of statements. While some work on phenomena like subjectivity has been carried out in the context of sentiment processing, other phenomena like the detection of sarcasm have received less attention.
By proposing this workshop we aim at bringing together scientists working on EPAM from any area related to computational language learning and processing. By EPAM we understand aspects of meaning that cannot be captured with a propositional representation such as the output of semantic role labelers.
For instance, the meaning of the sentence in Example (1) can be represented with the proposition ADD(earthquake,further threats to the global economy), whereas representing the meaning of the sentences in Example (2) requires additional mechanisms, despite the fact that all sentences share a propositional meaning.
(1) The earthquake adds further threats to the global economy.
(2) Does the earthquake add further threats to the global economy?
The earthquake adds further threats to the global economy, doesn’t it?
The earthquake does not add further threats to the global economy.
The earthquake will never add further threats to the global economy.
The earthquake will probably add further threats to the global economy.
Who could (possibly) think the earthquake adds further threats to the global economy?
The earthquake might have added further threats to the global economy.
The last analysis shows that the earthquake will add further threats to the global economy.
It is expected that the earthquake will add further threats to the global economy.
It has been denied that the earthquake adds further threats to the global economy.
Some of the sentences above could also be combined in a paragraph such as (3), which shows that the same event can be presented from different perspectives, at different points in time and with different extra-propositional meanings.
(3) The main question 6 months ago was whether the earthquake would add further threats to the global economy. Some days after the earthquake the authorities were convinced that it would be possible to minimize the impact of the earthquake. Most economists didn’t share this view and predicted a high economic impact of the earthquake. However, a recent study about the earthquake’s effect has shown that, although the earthquake might have added further threats to the global economy, its negative impact can be controlled by applying the right measures.
While the area of EPAM comprises a broad range of phenomena, this workshop will focus mainly on the aspects related to modality understood in a general sense (modalities, hedging, certainty, factuality), negation, attitude, and irony/sarcasm. Since many of these phenomena cannot be adequately modeled without taking (discourse) context into account, the workshop also touches on discourse phenomena in so far as they relate to extra-propositional aspects of meaning.
The workshop is a follow-up to Negation and Speculation in Natural Language Processing (NeSp-NLP 2010) held in Uppsala, Sweden, in July 2010.
SCOPE AND TOPICS
In particular, the workshop will address the following topics, although it will be open to other related topics:
– Negation
– Modality
– Hedging
– Factuality
– Certainty
– Subjectivity, attitude
– Evidentiality
– Irony, sarcasm
– Modeling and annotating extra-propositional aspects of meaning
– Scope resolution
– Detection of non-factual information
– Changes of the factual status of events within a text/message and within collections of texts/messages
– Discourse phenomena related to extra-propositional aspects of meaning
– The impact of extra-propositional aspects of meaning in NLP tasks: sentiment analysis, text mining, textual entailment, information extraction, machine translation, paraphrasing
– Implicit expression of extra-propositional meaning
– Multimodal expression of extra-propositional meaning
– Author profiling based on extra-propositional aspects of meaning
– Extra-propositional aspects of meaning across domains and genres
SUBMISSIONS
Authors are invited to submit full papers on original, unpublished work in the topic area of this workshop. All submissions must conform to the official ACL 2012 style guidelines and should not exceed 8 pages. Formatting instructions and the ACL 2012 Style Files can be found at http://www.acl2012.org/call/sub01.asp .
The reviewing of the papers will be blind and the papers should not include the authors’ names and affiliations. Each submission will be reviewed by at least three members of the program committee. Accepted papers will be published in the workshop proceedings.
Papers should be submitted no later than March 18, 2012, via the following submission site:
https://www.softconf.com/acl2012/exprom-2012
IMPORTANT DATES
March 18, 2012 – Submission deadline
April 15, 2012 – Notification of acceptance
April 30, 2012 – Camera-ready papers due
July 12, 13, or 14, 2012 – Workshop
ORGANISATION
Roser Morante, CLiPS-LTG, University of Antwerp
Caroline Sporleder, MMCI / Computational Linguistics and Phonetics, Saarland University
With financial support of the PASCAL2 Network.
PROGRAM COMMITTEE
Johan Bos – University of Groningen
Gosse Bouma – University of Groningen
Walter Daelemans – University of Antwerp
Roxana Girju – University of Illinois at Urbana-Champaign
Iris Hendrickx – University of Lisbon
Halil Kilicoglu – Concordia University
Maria Liakata – University of Wales
Katja Markert – University of Leeds
Erwin Marsi – Norwegian University of Science and Technology
David Martínez – NICTA and University of Melbourne
Malvina Nissim – University of Bologna
Sebastian Padó – University of Heidelberg
Sampo Pyysalo – NaCTeM and University of Manchester
Owen Rambow – Columbia University
Paolo Rosso – Universidad Politécnica de Valencia
Josef Ruppenhofer – Saarland University
Roser Saurí – Barcelona Media Innovation Center
Carlo Strapparava – Fondazione Bruno Kessler
György Szarvas – TU Darmstadt
Erik Velldal – University of Oslo
Annita de Waard – Elsevier Labs
Bonnie Webber – University of Edinburgh
Michael Wiegand – Saarland University
Call for Abstracts for a PASCAL sponsored event:
* Machine Learning and Applications stream *
EURO 2012, 25th European Conference on Operations Research,
July 8-11, 2012, Vilnius, Lithuania
(www.euro-2012.lt)
EURO is a major international conference in optimisation
which consists of several streams. We are organising the
*Machine Learning and Applications stream*,
and two sessions within it:
– Computer Vision
– Ensemble Learning and Ensemble Pruning Methods
This stream will enable Machine Learning researchers
to present their work to an audience of people interested in
machine learning and optimisation.
Abstracts (up to 600 characters) are welcome in any area of
machine learning, in particular works presenting applications
of optimisation.
Selected authors will be invited to present their work in a 20 min talk.
* Award for PASCAL members *
The best abstracts submitted by members of the PASCAL network
will be selected for sponsorship of all the travel expenses.
If you would like to be considered for this selection,
please contact us.
* Instructions *
In order to submit a contribution, please go to the website below and sign up:
http://www.euro-online.org/conf/display.php?page=welcome
Then, if you would like to take part in the Computer Vision track,
please paste the following code 2252d763 in the form.
Alternatively, if you would like to take part in the Ensemble Learning track, please use
the code 0172f368.
Next, click on “Submit invited abstract” and follow the instructions.
Best regards,
Teo de Campos, Fei Yan, Sureyya Ozogur-Akyuz and Terry Windeatt
Machine learning, probabilistic modelling, data mining etc. researchers are welcome!
Helsinki Institute for Information Technology HIIT (http://www.hiit.fi/) invites applications for
Postdoctoral researchers
Excellent applicants in the research fields of HIIT are welcome and the positions will be filled for three years maximum, starting 2 May 2012, or as agreed. For more detailed information, please see the full call text at http://www.hiit.fi/node/1465. The closing date of the call is February 29, 2012.
———————————————————————————————————
Helsinki Institute for Information Technology HIIT conducts basic and strategic research on information technology. It is a joint research institute of the Aalto University and the University of Helsinki. HIIT is a leading IT institute with research ranging from fundamental methods and technologies to novel applications and their impact on other sciences, people and society. HIIT’s key competences are in analysis of large sets of data, probabilistic modeling of complex phenomena, Internet architecture and technologies, mobile and
human-centric computing, and user-created media.
Several research fellowships available, Looking for researchers of exceptional talent to conduct 2 year research fellowship to work closely with Professor Philip Torr (http://cms.brookes.ac.uk/staff/PhilipTorr/), at Oxford Brookes vision research group http://cms.brookes.ac.uk/research/visiongroup to work on various projects connected with our ongoing work on scene understanding and natural language description of scenes (see here for recent publications http://cms.brookes.ac.uk/staff/PhilipTorr/papers.htm).
There will be a large amount of freedom for the researcher and it is expected that applicants will have a top flight record, as evinced by publications in top venues such as ICCV, NIPS, ECCV, SIGGRAPH, CVPR, PAMI, JMLR, IJCV etc. The candidate will benefit from close mentoring to encourage them to develop into a top flight researcher.
The computer vision group at Oxford Brookes has a strong international reputation, having taken scientific best paper awards at all the conferences mentioned previously, it is an exceptionally stimulating environment, with strong connections to the top industrial research labs such as Sony, Microsoft and Google, as well as very close links with the Visual Geometry Group at Oxford University with whom we share reading groups and seminars. It is soon to move to brand new facilities near the heart of Oxford, a very academically vibrant city.
For further information contact philiptorr@brookes.ac.uk (http://cms.brookes.ac.uk/staff/PhilipTorr/)
Topic: Adaptive user interfaces
User interfaces that adapt to the user’s activities, situation and knowledge can significantly improve interaction. This is particularly true of mobile devices and the next generation of smart phones are likely to have highly adaptive user interfaces. Achieving such adaptation requires techniques for learning from interaction data and modelling the user. This PhD will investigate those techniques and will develop novel adaptive user interfaces for mobile platforms.
The student will be supervised by Dr John Dowell and based in the Computer Science department within the user interaction research centre UCLIC. The research will involve a substantial collaboration with Orange Labs in west London. The studentship will provide an enhanced and tax free stipend of £18,120.
We are looking for someone who has:
• an excellent undergraduate degree in Computer Science
• an interest in user interfaces and user interaction and an interest in working with user interaction data, including running trials with users
• knowledge of computational techniques useful for adaptive systems such as agent modelling and machine learning; strong programming skills are required
• an interest in working within a multi disciplinary research centre
How to apply:
To be considered, you must fill in the general UCL application form. Please see http://www.ucl.ac.uk/prospective-students/graduate-study/application-admission/ where you can download the forms and guidelines. Make sure you specify the supervisor as John Dowell and ‘Adaptive user interfaces’ as the research subject area. The post will be filled as soon as a suitable candidate is found. We encourage those interested to apply as soon as possible.
Start date: before October 2012
Eligibility requirements: Standard EPSRC eligibility requirements hold, so to apply you must be either a UK resident or an EU citizen having completed your B.Sc. degree at a UK university.
Multivariate and predictive analysis of neuroimaging data has gained ground
very rapidly in the community over the past few years, leading to impressive
results in cognitive, affective, and clinical neurosciences. Innovations in
machine learning, such as mixed-norm regularisation, multiple kernel learning,
and online learning have been incorporated swiftly, and novel methods are
emerging which are specifically tuned to the constraints of neuroimaging data,
prompting advances in areas such as structured sparsity or covariate
modelling. Pattern recognition and machine learning conferences now typically
feature a neuroimaging workshop, while neuroscience and brain imaging
meetings dedicate sessions and track to “brain decoding” and multivariate
predictive methods. Thus, a rich two-way flow has been established between
disciplines.
After Istanbul (Workshop on Brain Decoding 2010) and Seoul
(PRNI 2011), it is the intention of the 2nd International Workshop on Pattern Recognition in NeuroImaging to continue facilitating exchange of ideas between scientific communities, with a particular interest in the link between mass-univariate, post-hoc modelling and multivariate predictive models.
** Topics of interest
PRNI welcomes original papers on multivariate predictive models of
neuroimaging data, using e.g. fMRI, sMRI, EEG, MEG, ECoG modalities,
including but not limited to the following topics:
* Learning from neuroimaging data
Online, incremental, and adaptive learning
Modality combinations
Optimisation and regularisation
Graph-based techniques and graphical models
* Interpretability of models and results
High-dimensional data visualisation
Multivariate and multiple hypothesis testing
Links between brain structure and function
Summarisation / presentation of inference results
* Applications
Disease diagnosis and prognosis
Real-time fMRI
Resting-state modelling
Cognitive neurosciences
** Submission Guidelines and Proceedings
Authors should prepare full papers with a maximum length of 4 pages (double-column, IEEE style, PDF) for review. Proceedings will be published by IEEE Computer Science Society in electronic format. They will be permanently available on the IEEExplore and IEEE CS Digital Library online repositories, and indexed in IET INSPEC, EI Compendex (Elsevier), Thomson ISI, and others. Participants will receive a CDROM. The workshop website has all the details:
http://www.mlnl.cs.ucl.ac.uk/prni2012/
** Important Dates
Paper submission deadline: 1st of April, 2012
Acceptance notification: 7th of May, 2012
Workshop: July 2-4, 2012**
=======================================================
For more details see: http://www.peterjbentley.com/heartworkshop/
Co-located with AISTATS 2012, the Classifying Heart Sounds Workshop is the first international workshop to focus on the use of statistical machine learning techniques to segment and classify real-world heart audio. This exciting one-day event will feature leading experts on auscultation, signal processing and machine learning, and will include presentations by researchers who have attempted the Classifying Heart Sounds Challenge, see below.
Free Registration! Attendance at the workshop is free, however we only have a limited amount of space, so please register if you plan to attend so that we can guarantee your place. To register, please send an email to Yiqi Deng
Reminder of the PASCAL Heart Sounds Challenge
=============================================
The PASCAL-sponsored Heart Sounds Challenge is your chance to prove your machine learning technique can outperform those of everyone else – and win an iPad for your efforts! (Also come to the Canary Islands to present your results in a workshop after AISTATS!)
For more details see: http://www.peterjbentley.com/heartchallenge/
According to the World Health Organisation, cardiovascular diseases (CVDs) are the number one cause of death globally: more people die annually from CVDs than from any other cause. An estimated 17.1 million people died from CVDs in 2004, representing 29% of all global deaths. Of these deaths, an estimated 7.2 million were due to coronary heart disease. Any method which can help to detect signs of heart disease could therefore have a significant impact on world health. This challenge is to produce methods to do exactly that. Specifically, we are interested in creating the first level of screening of cardiac pathologies both in a Hospital environment by a doctor (using a digital stethoscope) and at home by the patient (using a mobile device).
For this challenge we have two datasets comprising several hundred real heart sounds, gathered from an iphone app by the general public, and by a digital stethoscope in a noisy hospital environment.
Challenge 1 is segmentation – can your method correctly identify the “lub dub” (S1 and S2) components of the sound?
Challenge 2 is classification – can your method correctly classify the heart sounds into categories such as Normal, Murmur, Extra Heart Sound, and Artifact?
This problem is of particular interest to machine learning researchers as it involves classification of audio sample data, where distinguishing between classes of interest is non-trivial. Data is gathered in real-world situations and frequently contains background noise of every conceivable type. The differences between heart sounds corresponding to different heart symptoms can also be extremely subtle and challenging to separate. Success in classifying this form of data requires extremely robust classifiers. Despite its medical significance, to date this is a relatively unexplored application for machine learning.
Enquiries and submission, email: Yiqi Deng
