Associate Research Fellow – University of Exeter

University of Exeter – College of Engineering, Mathematics and
Physical Sciences

The College wishes to recruit an Associate Research Fellow to support
the work of Dr Yiming Ying. This EPSRC funded post is available from 1
January 2012 on a fixed term basis for 16 months until 30 April 2013.
The successful applicant will contribute towards research related to
the project entitled ‘Towards a new generation of matrix learning
methods in machine learning’.

The successful applicant will undertake computational and theoretical
work under this project. In particular, she/he is expected to develop
new ML models and algorithms, analyze optimisation algorithms and
extensively conduct empirical studies on large and challenging
application problems such as face identification and collaborative
filtering. The successful applicant will be able to present
information on research progress and outcomes, communicate complex
information, orally, in writing and electronically.

The starting salary will be from £24,370 up to £26,063 on Grade E,
depending on qualifications and experience.
The application deadline is 31/10/2011

For further information please contact Dr. Yiming Ying, e-mail
y.y…(at)ex.ac.uk or telephone (01392) 723591.
To view further details, please go to http://empslocal.ex.ac.uk/people/staff/yy267/

CFP / ACM TIST Special Issue on Social Web Mining

Call for Papers
ACM Transactions on Intelligent Systems and Technology (ACM TIST)
Special Issue on Social Web Mining

Social network platforms (e.g., Facebook, Google+, Twitter) have been
drawing increasing attention from everyday users worldwide to research
communities and industries. For instance, it is reported that there
are more than 750 million active Facebook users, of which 50% log on
in any given day. The success of the social network platform provides
a wealth of information, which brings great opportunities for
e-Commerce, policy making, social media recommendations, and much more.
However, it can be challenging to model the social networks and make
predictions, given this prosperity of social networks leading to
information overload.

Social web mining aims at modeling social interactions (e.g., posting
and messaging in Facebook) and relationships (e.g., friendship in
Facebook, Followed / Following in Twitter, Friends / Family /
Acquaintances / Following in Google+), and then learning inductive
rules for predictions, which are carried out by leveraging the
information available via various social networks and recent advances
in machine learning and data mining. This special issue is aimed at
providing a forum to highlight the leading works from the data mining
and machine learning communities for making sense of social networks.

Topics of interest include but are not limited to:
– Community detection
– Crawling of social networks
– Large scale graph mining
– Multi-relational data mining
– News/social media recommender systems
– Preference learning and elicitation in social networks
– Privacy protection in social networks
– Sentiment analysis in social networks (e.g., opinion mining in Twitter)

Submissions

On-Line Submission (will be available around January 1, 2012):
http://mc.manuscriptcentral.com/tist (please select “Special Issue:
Social Web Mining” as the manuscript type)
Details of the journal and manuscript preparation are available
on the website http://tist.acm.org/
Each paper will be peer-reviewed by at least three reviewers.

Important Dates

Full Paper Submission Deadline: February 1, 2012
Review Notification: April 1, 2012,
Final Manuscript: May 1, 2012,
Electronic Publication Date: June, 2012,

Guest Editors

Francesco Bonchi
Yahoo! Research Barcelona, Spain
http://www.francescobonchi.com/

Wray Buntine
NICTA – ANU, Australia
http://www.nicta.com.au/people/buntinew

Ricard Gavaldà
Technical University of Catalonia, Spain
http://www.lsi.upc.edu/~gavalda/

Shengbo Guo
Xerox Research Centre Europe, France
http://www.xrce.xerox.com/Research-Development/Services-Innovation-Laboratory/Machine-Learning-for-Optimisation-and-Services/People/Shengbo-Guo

Questions?
bonchi(at)yahoo-inc.com, wray.buntine(at)nicta.com.au,
gavalda(at)lsi.upc.edu, shengbo.guo(at)xrce.xerox.com

COLT2012 Call for Papers

The 25th Annual Conference on Learning Theory (COLT 2012) will take place
in Edinburgh, Scotland, on June 25th-27th, 2012. It will be co-located with
ICML, with a joint day on June 27th.

We invite submissions of papers addressing theoretical aspects of machine
learning, empirical inference and related topics. We strongly support a
broad definition of learning theory, including:

* Analysis of learning algorithms and their generalization ability
* Computational complexity of learning
* Statistical mechanics of learning systems
* Optimization procedures for learning and their analysis
* Online and stochastic optimization, and analysis of oracle models of optimization
* High dimensional and non-parametric empirical inference, including sparse and low-rank
methods
* Kernel methods
* Boolean function learning
* Unsupervised and semi-supervised learning and clustering
* On-line learning and relative loss bounds
* Planning and control, including reinforcement learning
* Learning in social, economic, and game-theoretic settings
* Privacy in learning

We are also interested in papers that include viewpoints that are new to
the COLT community. We welcome experimental and algorithmic papers provided
they are relevant to the focus of the conference by elucidating theoretical
results, or by pointing out interesting and not well understood behavior
that could stimulate theoretical analysis.

Papers that have previously appeared in journals or at other conferences
with published proceedings, or that are being submitted to other
conferences with published proceedings, are not appropriate for COLT.

Papers that include work that has already been submitted for journal
publication may be submitted to COLT, as long as the papers have not been
accepted for publication by the COLT submission deadline (conditionally or
otherwise) and that the paper is not expected to be published before the
COLT conference (June 2012).

Papers will be published electronically in the JMLR Workshop and
Conference Proceedings series.

Open Problems Session:
We also invite submission of open problems. These should be constrained to
two pages. There is a short reviewing period for the open problems.

Accepted contributions will be allocated short presentation slots in a
special open problems session and will be allowed two pages each in the
proceedings. The deadline for submitting open problems will be announced
in a separate call at a later date.

Submission Instructions:
Submissions are limited to 12 JMLR-formatted pages, plus additional pages
for references and appendices. All details, proofs and derivations
required to substantiate the results must be included in the submission,
possibly in the appendices. However, the contribution, novelty and
significance of submissions will be judged primarily based on the main text
(without appendices), and so enough details, including proof details, must
be provided in the main text to convince the reviewers of the submissions
merits.

Exact formatting and submission instructions will be available at the
conference website http://www.ttic.edu/colt2012/

Important Dates:
Paper submission deadline: Tuesday February 14th, 2012, 18:00 UTC (1PM EST).
Author notification: April 30th, 2012
Conference: June 25th-27th, 2012

General Chair: Jykri Kivinen
Open Problem Chair: Alexander Rakhlin
Program Chairs: Shie Mannor and Nathan Srebro

Program Committee:

Peter Auer
Francis Bach
Maria-Florina Balcan
Shai Ben-David
Kamalika Chaudhuri
Koby Crammer
Vitaly Feldman
Claudio Gentile
Peter Grunwald
Elad Hazan
Adam Kalai
Satyen Kale
Robert Kleinberg
Vladimir Koltchinskii
John Lafferty
Phil Long
Gabor Lugosi
Yishay Mansour
Remi Munos
Pardeep Ravikumar
Philippe Rigollet
Shai Shalev-Shwartz
Ohad Shamir
Alexander Shapiro
Ingo Steinwart
Gilles Stoltz
Csaba Szepesvari
Manfred Warmuth
Hans Simon

Postdoctoral Position: Distance Learning For Person Recognition

The Computer Vision group of the University of Caen (France) and the LEAR group at INRIA Grenoble, Rhone-Alpes (France) are opening a postdoctoral position. The candidate will be jointly supervised and will spend time in both institutions (http://lear.inrialpes.fr/; http://www.unicaen.fr/).

Topic: person recognition and re-identification across networks of cameras is an important problem in computer vision and is required for a large number of applications (such as video surveillance). However, despite the effort of the community, person re-identification is still a challenging problem: the same person can look very differently according to his posture or the camera position, while the difference between two different persons can be small. The goal of this project is to develop algorithms for finding occurrences of the same persons in huge collections of video surveillance images. The key issue is to learn distance functions from the training data, which embed visual information. This research will build on papers such as “Is that you? Metric learning approaches for face identification”, G. Matthieu, V. Jakob and Schmid Cordelia, ICCV 2009” and “Randomized Clustering Forests for Image Classification”, F Moosmann, E Nowak and F Jurie, PAMI 2008.

Start date: As soon as possible.

Location: The student will be located at the Computer Vision group at the University of Caen, and will spend approximately a 1 week every 2 months at the INRIA Grenoble research center.

Contacts:
Prof. Frederic Jurie, frederic.jurie(at)unicaen.fr
Res. Dir. Cordelia Schmid, cordelia.schmid(at)inrialpes.fr

Profile:
* Recent PhD (preferably in Computer Science, computer vision, machine learning or Applied Mathematics)
* Solid programming skills; the project involves programming in Matlab and C++
* Solid mathematics knowledge (especially linear algebra and tatistics)
* Creative and highly motivated
* Fluent in English, both written and spoken
* Prior knowledge in the areas of computer vision, machine learning or data mining is a plus (ideally a Master thesis in a related field)

Please send applications via email, including:
* a complete CV including a list of publications
* graduation marks as well as rank
* the name and email address of three references

Applications should be sent to Frederic Jurie (frederic.jurie(at)unicaen.fr).

Applicants can be asked to do a short assignment in order to demonstrate their research abilities.

PhD studentship – University of Sheffield

EPSRC PhD Studentship
University of Sheffield – Department of Automatic Control and Systems
Engineering

Project title: Real-time Ray Tracing

Supervisor: Professor Dave Allerton and Professor Visakan
Kadirkamanathan
Start date: 9th January 2012 (3 years)

Full details available at http://www.pascal-network.org/docs/phdshef.pdf

Call for Papers – KDD2012

CALL FOR PAPERS
18th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD-2012)
August 12-16, 2012
Beijing, China

http://www.kdd.org/kdd2012/

Key Dates:
Papers due: February 10, 2012
Acceptance notification: May 4, 2012

Paper submission and reviewing will be handled electronically. Authors should consult the conference Web site for full details regarding paper preparation and submission guidelines.

Papers submitted to KDD 2012 should be original work and substantively different from papers that have been previously published or are under review in a journal or another conference/workshop.

As per KDD tradition, reviews are not double-blind, and author names and affiliations should be listed.

Due to the large number of submissions, papers submitted to the research track will not be considered for publication in the industry/government track and vice-versa. Authors are encouraged to carefully read the conference CFP and choose an appropriate track for their submissions. In case of doubts, authors are encouraged to get in touch with the chairs of the corresponding track at least a week before the submission deadline.

RESEARCH TRACK

We invite submission of papers describing innovative research on all aspects of knowledge discovery and data mining. Examples of topic of interest include (but are not limited to): association analysis, classification and regression methods, semi-supervised learning, clustering, factorization, transfer and multi-task learning, feature selection, social networks, mining of graph data, temporal and spatial data analysis, scalability, privacy, security, visualization, text analysis, Web mining, mining mobile data, recommender systems, bioinformatics, e-commerce, online advertising, anomaly detection, and knowledge discovery from big data, including the data on the cloud. Papers emphasizing theoretical foundations, novel modeling and algorithmic approaches to specific data mining problems in scientific, business, medical, and engineering applications are particularly encouraged. We welcome submissions by authors who are new to the KDD conference, as well as visionary papers on new and emerging topics. Authors are explicitly discouraged from submitting papers that contain only incremental results and that do not provide significant advances over existing approaches. Application oriented papers that make innovative technical contributions to research are welcome.

Submitted papers will be assessed based on their novelty, technical quality, potential impact, and clarity of writing. For papers that rely heavily on empirical evaluations, the experimental methods and results should be clear, well executed, and repeatable. Authors are strongly encouraged to make data and code publicly available whenever possible.

INDUSTRY & GOVERNMENT TRACK

The Industrial/Government Applications Track solicits papers describing implementations of KDD solutions relevant to industrial or government settings. The primary emphasis is on papers that advance the understanding of practical, applied, or pragmatic issues related to the use of KDD technologies in industry and government and highlight new research challenges arising from attempts to create such real KDD applications. Applications can be in any field including, but not limited to: e-commerce, medical and pharmaceutical, defense, public policy, finance, engineering, environment, manufacturing, telecommunications, and government.

The Industrial/Government Applications Track will consist of competitively-selected contributed papers. Submitters must clearly identify in which of the following three sub-areas their paper should be evaluated as distinct review criteria will be used to evaluate each category of submission.

· Deployed KDD systems that are providing real value to industry, Government, or other organizations or professions. These deployed systems could support ongoing knowledge discovery or could be applications that employ discovered knowledge, or some combination of the two.

· Discoveries of knowledge with demonstrable value to Industry, Government, or other users (e.g., scientific or medical professions). This knowledge must be “externally validated” as interesting and useful; it can not simply be a model that has better performance on some traditional KDD metric such as accuracy or area under the curve.

· Emerging applications and technology that provide insight relevant to the above value propositions. These emerging applications must have clear user interest and support to distinguish them from KDD research papers, or they must provide insight into issues and factors that affect the successful use of KDD technology and methods. Papers that describe infrastructure that enables the large-scale deployment of KDD techniques also are in this area.

ON BEHALF OF THE KDD-2012 ORGANIZERS

Research Program Co-chairs:

· Deepak Agarwal, Yahoo! Research

· Jian Pei, Simon Fraser University

Industry and Government Program Co-chairs:

· Michael Zeller, Zementis

· Hui Xiong, Rutgers University

General Chair:

· Qiang Yang, HKUST

Associate General Chair

· Dou Shen, CityGrid Media

ICMI 2012 Call for Multimodal Grand Challenges

International Conference on Multimodal Interaction (ICMI) 2012
Santa Monica, CA, Oct. 22-26, 2012
First Call for Multimodal Grand Challenges
—————————————————————–

The International Conference on Multimodal Interaction (ICMI) is the
premier international forum for multidisciplinary research on multimodal
human-human and human-computer interaction, interfaces, and system development. The
conference focuses on theoretical and empirical foundations, component
technologies, and combined multimodal processing techniques that define the
field of multimodal interaction analysis, interface design, and system
development.

Multimodal technologies are fundamentally changing the ways people interact
with each other and with computers. Multimodal systems – recognizing
language and gesture exchanged among people or as input to interfaces in physical
environments or mobile devices – facilitate and enrich social
interactions and allow people to naturally interact with computer systems with minimal
training, for domains such as communications, education, entertainment, and robotics.

Developing systems that can robustly understand human-human
communication or respond to human input requires identifying the best algorithms and their
failure modes. In fields such as computer vision, speech recognition, and
computational linguistics, the availability of datasets and common tasks have
led to great progress. We invite the research community to collectively define
and tackle the scientific Grand Challenges in our domain for the next 5 years.
Multimodal Grand Challenges are driven by ideas that are bold,innovative, and
inclusive. They should inspire new ideas in the multimodal interaction
community and create momentum for future collaborative work. Analysis,
synthesis, and interactive tasks are all possible.

We are seeking organizers to propose and run Grand Challenge events. Both
academic and corporate organizers are welcome. We are looking for three
types of challenges:

1. Dataset-driven challenge. This challenge will provide a dataset that is
exemplary of the complexities of current and future multimodal problems,
and one or more multimodal tasks whose performance can be objectively measured.
Participants in the Challenge will evaluate their methods against the
challenge data in order to identify areas of strength and weakness.

2. User case challenge. This challenge will provide an interactive
problem or system (e.g. dialog-based) and the associated resources,
which allow people to participate through the integration of specific
modules or alternative full systems. Proposers should also establish
systematic evaluation procedures.

3. Concept challenge. This challenge proposes new ideas (e.g. involving new
sensors) that, while not fully tested now, could lead to breakthroughs
if the community decided to tackle them together or individually.

Prospective organizers should submit a five-page max proposal containing
the following information:

– Title
– Abstract appropriate for possible Web promotion of the Challenge
– Detailed description of the challenge and its relevance to multimodal
interaction.
– Plan for soliciting participation
– Proposed schedule for releasing datasets and receiving submissions.
– Short bio of the organizers
– Funding source (if any) that supports or could support the challenge
organization.
– Preference (if any) for special session or workshop format.

Proposals will be evaluated based on originality, ambition, feasibility,
and implementation plan. The ICMI organizers offer support with basic
logistics and with the identification of additional funding sources (e.g. for awards).

Proposals should be emailed to the ICMI 2012 Challenge Co-Chairs.
Prospective organizers are also encouraged to contact them in case of questions.

Important dates

Proposal submissions: December 9, 2011
Notifications: December 16, 2011
ICMI 2012: October 22-26, 2012

ICMI 2012 Grand Challenge Chairs

Daniel Gatica-Perez
Idiap Research Institute
gatica(at)idiap.ch

Stefanie Tellex
MIT
stefie10(at)csail.mit.edu

CFP – CVMP (Conference for Visual Media Production) 2011 – EXTENDED DEADLINE

16th and 17th November 2011
BFI SouthBank, London
www.cvmp-conference.org

Submission deadline: *Extended to 30 June 2011*

Full call details available at http://www.pascal-network.org/docs/cvmp11.pdf

NIPS 2011 Registration Is Open

NIPS 2011 Registration is now open:
https://nips.cc/Register/

Early registration pricing expires after November 6, 2011

This year the tutorials and conference sessions are being held at the
Palacio de Exposiciones y Congresos (Congress Center) in Granada, Spain

The workshops are being held in the Sierra Nevada, just outside Granada.

Program highlights may be found here:
http://nips.cc/Conferences/2011/Program/

Because the conference is in a new country there have been important
changes. Please read more at this URL:
http://nips.cc/ConferenceInformation/ChangeLog

For travel support information:
http://nips.cc/ConferenceInformation/TravelSupport

For volunteering information:
http://nips.cc/ConferenceInformation/Volunteering

Additional conference information can be found here:
http://nips.cc/ConferenceInformation/

We look forward to seeing you in Spain!

KDD-2012 CALL FOR WORKSHOPS

[General Information]

The ACM KDD-2012 organizing committee solicits proposals for workshops
to be held in conjunction with the main conference. Each workshop will
take place on August 12, 2012 (tentatively) and will either be a
full-day or a half-day workshop. The purpose of a workshop is to provide
an excellent opportunity for participants from academia, industry,
government and other related parties to present and discuss novel ideas
on current and emerging topics relevant to knowledge discovery and data
mining.

Each workshop should be organized under a well-defined basis focusing on
emerging research areas, challenging problems and
industrial/governmental applications. Organizers have free controls on
the format, style as well as building blocks of the workshop. Possible
contents of a workshop include but are not limited to invited talks,
regular papers/posters, panels, and other pragmatic alternatives. In
case workshop proposers need extra time to prepare their workshop, early
decisions may be considered if justified.

Organizers of accepted workshops are expected to announce the workshop
and disseminate call for papers, maintain the workshop website, gather
submissions, conduct the reviewing process and decide upon the final
workshop program. They are also required to prepare an informal set of
workshop proceedings to be distributed with the registration materials
at the main conference, with a proceedings format template provided by
KDD 2012. Workshop organizers may choose to form organizing or program
committees aiming to accomplish these tasks successfully.

[Important Dates]

· Workshop proposals due: January 15, 2012

· Notification of decision: March 5, 2012

· Suggested Workshop Paper Submission deadline: May 8, 2012

· Suggested Workshop Final Paper Due: June 18, 2012

· Workshop Proceedings Due: June 25, 2012

[Proposal Details]

Workshop proposals should be no more than 8 pages and must include the
following parts:

· Outline: abstract, topic(s), objectives, relevance, and expected
outcomes

· Motivation: why is SIGKDD workshop on this topic relevant at
this particular time

· Description of the target group(s) of attendees and anticipated
number of participants

· Potential list of invited speakers (if any)

· Preliminary list of core program committee members

· Duration of the workshop (full-day or half-day)

· For workshops previously held at KDD or other conferences,
details on venue, attendance and number of submissions/accepted papers
from past editions

· For new workshops, a list of possible attendees/submissions
and/or a justification of the expected attendees/submissions

· Short bio as well as contact information (address, email, and
phone) for each organizer

· A designated contact person

Proposers are encouraged to have their drafts reviewed by potential
workshop participants before submission.

Workshop proposals should be submitted by January 15, 2011 at the
following url:

https://www.easychair.org/conferences/?conf=kdd2012ws

[Notes]

The ACM KDD-2012 organizing committee grants each accepted workshop with
one FREE conference registration, which can be offered to one of the
workshop organizers themselves or the workshop invited speakers. The
workshops are also encouraged to seek their own sponsors. Workshop
organizers are not allowed to publish more than two papers in their own
workshop. At least one author of each accepted paper is required to
register to the workshop and present their work.

[Workshop Co-Chairs]

· Zhi-Hua Zhou, Nanjing University

· Sofus A. Macskassy, ISI/USC

Contact info: workshops(at)kdd2012.com