News Archives

Machine Learning Postdoc position

Research Area: Low-Rank Matrix Recovery and Approximation, Sparse Coding.

Project Description: Applications are invited for an open Postdoctoral
Research Scientist position at SUNY at Buffalo, Department of
Computer Science and Engineering, in the area of
machine learning. Qualified candidates must have a Ph.D. in
machine learning or related areas with
outstanding research record and experience. The grant support will be
3 years. Successful candidates will conduct basic research and
interact with the principal investigator, graduate students, and
collaborators. The Computer Science department at SUNY Buffalo is among
the oldest CS departments nationwide with a strong focus on computer
vision and machine learning.
See http://www.cse.buffalo.edu/ for more information.

Salary is sufficiently competitive. If you are interested in joining
this research project as a Postdoctoral Fellow, please contact:

Yun (Raymond) Fu, Principal Investigator
Department of Computer Science and Engineering
State University of New York (SUNY) at Buffalo
201 Bell Hall Buffalo, NY 14260-2000, USA
Ph: +1 (716) 645 2670
Email: yunfu(at)buffalo.edu
Web: http://www.cse.buffalo.edu/~yunfu/

Travel Support: New Frontiers in Model Order Selection — NIPS-2011 Workshop — Call for Abstracts

New Frontiers in Model Order Selection
NIPS-2011 Workshop, Granada, Spain,
December 16, 2011
http://people.kyb.tuebingen.mpg.de/seldin/fimos.html

UPDATE

New deadline: October 24!

We are glad to announce that we will provide travel support for workshop contributors (sponsored by PASCAL2). See the workshop page for details.

DESCRIPTION

Model order selection, which is a trade-off between model complexity and its empirical data fit, is one of the fundamental questions in machine learning. It was studied in detail in the context of supervised learning with i.i.d. samples, but received relatively little attention beyond this domain. The goal of our workshop is to raise attention to the question of model order selection in other domains, share ideas and approaches between the domains, and identify perspective directions for future research. Our interest covers ways of defining model complexity in different domains, examples of practical problems, where intelligent model order selection yields advantage over simplistic approaches, and new theoretical tools for analysis of model order selection. The domains of interest span over all problems that cannot be directly mapped to supervised learning with i.i.d. samples, including, but not limited to, reinforcement learning, active learning, learning with delayed, partial, or indirect feedback, and learning with submodular functions.

An example of first steps in defining complexity of models in reinforcement learning, applying trade-off between model complexity and empirical performance, and analyzing it can be found in [1-4]. An intriguing research direction coming out of these works is simultaneous analysis of exploration-exploitation and model order selection trade-offs. Such an analysis enables to design and analyze models that adapt their complexity as they continue to explore and observe new data. Potential practical applications of such models include contextual bandits (for example, in personalization of recommendations on the web [5]) and Markov decision processes.

References:
[1] N. Tishby, D. Polani. “Information Theory of Decisions and Actions”, Perception-Reason-Action Cycle: Models, Algorithms and Systems, 2010.
[2] J. Asmuth, L. Li, M. L. Littman, A. Nouri, D. Wingate, “A Bayesian Sampling Approach to Exploration in Reinforcement Learning”, UAI, 2009.
[3] N. Srinivas, A. Krause, S. M. Kakade, M. Seeger, “Gaussian Process Optimization in the Bandit Setting: No Regret and Experimental Design”, ICML, 2010.
[4] Y. Seldin, N. Cesa-Bianchi, F. Laviolette, P. Auer, J. Shawe-Taylor, J. Peters, “PAC-Bayesian Analysis of the Exploration-Exploitation Trade-off”, ICML-2011 workshop on online trading of exploration and exploitation.
[5] A. Beygelzimer, J. Langford, L. Li, L. Reyzin, R. Schapire, “Contextual Bandit Algorithms with Supervised Learning Guarantees”, AISTATS, 2011.

CALL FOR ABSTRACTS

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

IMPORTANT DATES

Submission Deadline: October 24, 8:00am GMT.
Notification of Acceptance: October 15.

INVITED SPEAKERS

Shie Mannor (tentative)
Sanjoy Dasgupta (tentative)
John Langford (tentative)
Naftali Tishby (tentative)
Peter Auer (tentative)

ORGANIZERS

Yevgeny Seldin, Max Planck Institute for Intelligent Systems
Koby Crammer, Technion
Nicolò Cesa-Bianchi, University of Milano
François Laviolette, Université Laval
John Shawe-Taylor, University College London

Research Associate in Natural Language Processing and Machine Learning

Faculty of Engineering

University of Sheffield – Department of Computer Science

Job Reference Number: UOS003396

Contract Type: Fixed-term for up to 36 months

Salary: Grade 7 £28,251 – £35,788 per annum.

Closing Date: 3 November 2011

Further details and online application: http://jobs.ac.uk/job/ADI311/

Summary:

The objective of this research project is to develop new machine learning techniques for predictive modelling of financial and political indices using text from social media sources (e.g., Twitter, Facebook and blogs). The project will develop algorithms for modelling the correlations between streaming social media data and the movement of various indices, viewed as a time-series. This problem presents unique challenges, both in terms of learning algorithms and in terms of efficient development for deployment in a real-time setting.

The appointee will work on the machine learning components of the project, namely developing new statistical models for our time-series data, and associated algorithms for training and prediction. The appointee will be responsible for developing new efficient algorithms for Bayesian inference. A central focus of the role will be developing fast online algorithms suitable for real-time application. The role will require strong programming skills, particularly for cluster and cloud computing infrastructure (e.g., MapReduce, Amazon EC2) or GPU computing (e.g., CUDA).

This is an opportunity to work in a well-connected international team with world-leading reputations in both the Natural Language Processing (NLP) and Machine Learning (ML) research groups at The University of Sheffield. The NLP group is well known internationally for its research, and is one of the largest research groups in computational linguistics and text engineering in the UK. The ML group is also very well respected, with expertise in fundamental machine learning and a range of application domains.

Candidates must have a PhD and a strong publication record in a relevant discipline. Solid knowledge of machine learning and natural language processing is required, as is excellent programming ability. The candidate should also have experience in one or more of the following areas: time-series modelling, dimensionality reduction/clustering, statistical machine translation, probabilistic graphical models, Markov Chain Monte Carlo and reinforcement learning.

This post is fixed-term for up to 36 months.

Researcher Position in Machine Learning for Neuroscience

Researcher Position in Machine Learning for Neuroscience is available in the Neuroinformatics Laboratory (NiLab) at Fondazione Bruno Kessler

The Neuroinformatics Laboratory (NILab) is a joint initiative between Fondazione Bruno Kessler (FBK) and the Center for Mind and Brain Sciences (CIMeC) of the University of Trento in order to promote interdisciplinary research in cognitive neuroscience. Neuroinformatics stands at the intersection of neuroscience and information science and it provides methods and technologies for managing, analyzing, and modeling neuroimaging data.

The NILab mission spans from scientific to technological aspects. The scientific research activity covers the design and the development of novel methods for the integration, the analysis and the interpretation of unimodal and multimodal neuroimaging data. The laboratory is co-located with the Neuroimaging Laboratory (LNIF) of CIMeC, which provides several facilities for cognitive neuroscience investigations such as MR (4T Brucker Medspec Scanner), MEG/EEG (Electa Neuromag), TMS and EyeTracking.

The successful candidate will work on the development of computational methods for brain data analysis approaching challenging tasks such as brain decoding, brain mapping and brain connectivity.

For further information, please contact info.nilab(at)fbk.eu.

Due to the FBK’s attempt to promote equal opportunity and gender balance, in case of equal applications, female candidates will be given preference.

The ideal candidate should have:
* Ph.D. in Computer Science or related fields.
* Solid background in Machine Learning and Pattern Recognition.
* Outstanding publication record.
* At least basic knowledge of the neuroimaging techniques (fMRI, dMRI, MEG, EEG).
* Good skills in scientific programming with Python.
* Proficient English both written and spoken.

Additional requirements / desiderata:
* Attitude to work in a multidisciplinary environment.
* Ability to quickly learn and use new technologies and tools.
* Ability to acquire knowledge from different application domains.

Type of Contract: research position for a 3 years contract starting November 2011. The gross salary offer will range between € 37,800.00 and € 45,000.00, depending on seniority and expertise.

Useful Links:
Neuroinformatics Laboratory – nilab.fbk.eu
Fondazione Bruno Kessler – www.fbk.eu
Center for Mind/Brain Sciences – www.cimec.unitn.it

To apply online please send your detailed CV with two references to jobs@fbk.eu.
Emails should have Ref.Code: NILAB_2011

[MLNI2011] Call for participation to the “Machine Learning for NeuroImaging Workshop”

This is to announce the “Machine Learning for Neuroimaging” Workshop that will be held on Nov. 8 and 9, 2011 in Marseille, France.

The main goal of this workshop is to bring together people from the machine learning community and people from the neuroimaging community that are keen to discuss their expertises. Potential outcomes to this workshop are for instance: the formal/machine learning setting of common problems in neuroimaging, the identification of new problems that can be readily tackled using machine learning techniques, the creation of new collaborations. It is also expected that discussions will build around important challenges of machine learning posed by neuroimaging data such as feature selection in presence of few data, transfer learning, structured prediction…

Further details can be found here:
http://mlni2011.sciencesconf.org/

** Important note ** : registration is free but limited to the first 40 participants; registration will be closed on Oct. 25, 2011.

Invited speakers: John Ashburner (FIL / UCL, London, UK), Francis Bach (Sierra / Inria, Paris, France), Olivier Colliot (Cogimage / CRICM, Paris, France), Edouard Duchesnay (LNAO / Neurospin, Gif sur Yvette, France), Thomas Gärtner (KDML / Universität Bonn, Bonn, Germany), Arthur Gretton (Gatsby / UCL, London, UK), Janaina Mourao-Miranda (UCL, London, UK), Alain Rakotomamonjy (LITIS / Université de Rouen, Rouen, France), Jonas Richiardi (EPFL, Lausanne, Switzerland), Marie Szafranski (ENSIIE, IBISC, Evry, France), Sylvain Takerkart (INCM / INT, Marseille, France), Bertrand Thirion (Parietal / Inria, Gif sur Yvette, France), Jean-Philippe Vert (Institut Curie / Mines ParisTech, Paris, France), Marcel Van Gerven (Donders Institute, Nijmegen, The Netherlands).

Looking forward to see you in Marseille

Xerox – Researcher, Statistical Machine Translation (2 positions)

The Machine Learning for Document Access and Translation group of the
Xerox Research Centre Europe conducts research in Statistical Machine
Translation and Information Retrieval, Categorization and Clustering
using advanced machine learning methods.

We are opening two positions for researchers with a background in
Statistical Machine Translation to support a combination of internal
and EU-funded projects.

Required experience and qualifications:

– PhD in computer science or computational linguistics with focus on
SMT or statistical NLP.
– Good publication record and evidence of implementing systems.
– A good command of English, as well as open-mindedness and the will
to collaborate within a team.

Preferred starting date: January 2012

Contract duration: 18 months

Application instructions

Please email your CV and covering letter, with message subject
“Statistical Machine Translation Researcher” to xrce-candidates and to
Nicola.Cancedda at xrce.xerox.com. Inquiries can be sent to
Nicola.Cancedda at xrce.xerox.com.

XRCE is a highly innovative place and we strongly encourage
publication and interaction with the scientific community.

Job announcement URLs:

http://www.xrce.xerox.com/About-XRCE/Career-opportunities/Researcher-Statistical-Machine-Translation

http://www.xrce.xerox.com/About-XRCE/Career-opportunities/Researcher-Statistical-Machine-Translation2

Postdoctoral scholar position in Machine Learning and Optimization at ETH Zurich

The newly established Learning and Adaptive Systems group at ETH
Zurich ( http://las.ethz.ch/ ), led by Andreas Krause, has an open
position for a postdoctoral scholar. The project involves large scale
active learning and sequential decision making based on
high-dimensional data, and will be carried out in collaboration with
researchers at the California Institute of Technology.

Applicants should have finished, or be about to finish their Ph.D.
degrees. They must have an exceptional background in machine learning
or optimization. Successful candidates need to have a strong track
record of publications at top machine learning, AI or theory
conferences (NIPS, ICML, COLT, AISTATS, AAAI, IJCAI, …) and/or
premier journals in the area (JMLR, JAIR, PAMI, …), and have
experience in at least one of the following areas:
– Active learning
– Online learning / bandits
– Combinatorial optimization / approximation algorithms
– Application of discrete optimization in ML and computer vision
– Sequential decision making under uncertainty / stochastic optimal control
– High dimensional statistics
– Algorithms for large scale probabilistic inference

The initial appointment is for 12 months, with possible extensions up
to 3 years.

Working language at ETH Zurich is English — German is not required.

The salary is highly competitive (among the highest in Europe).

Applicants are requested to send their
– CV incl. publication list
– two strongest publications relevant to this position
– contact information for three recommenders
to Andreas Krause ( krausea(at)ethz.ch ). Review of applications will
start immediately, and continue until the position is filled.

PhD student position in statistical machine learning, exploratory data analysis, visualization, and multi-task learning (Helsinki, Finland)

Aalto University School of Science invites applications for a

doctoral student / research assistant

position for a fixed term (initially 1 year, extension possible).
Start date: 14 November 2011 (negotiable).
Application deadline: 31 October 2011, 3pm Finnish local time.

You will develop advanced machine learning methods for nonlinear dimensionality reduction, visualization, and exploratory data analysis with multiple data sources. You will create novel probabilistic models for the structure and dependencies within and between multiple partly related data sets. The developed methods will be used in bioinformatics, neuroinformatics, analysis of structured data like graphs, and other applications. We have recently developed novel frameworks for visualization from an information retrieval perspective, and for multitask learning in asymmetric scenarios; your work will build on and extend these research lines.

Research Site:

The position is at the Department of Information and Computer Science, Aalto University School of Science (previously Helsinki University of Technology). The focus of the Department’s research and teaching activity is on advanced computational methods for modelling, analysing, and solving complex tasks in technology and science. The research aims at the development of fundamental computer science methods for the analysis of large and high-dimensional data sets, and for the modelling and design of complex software, networking and other computational systems. The department employs approximately 150 people and operates with a total annual budget of approximately 9 MEUR. The department hosts the Finnish Centre of Excellence in Adaptive Informatics Research (AIRC) and part of the Finnish Centre of Excellence in Algorithmic Data Analysis Research, and will host part of the new Finnish Centre of Excellence in Computational Inference Research (COIN) which starts in 2012. The depart!
ment was ranked among the top two departments of Aalto University in the Research Assessment Exercise 2009.

The position is located in the Statistical Machine Learning and Bioinformatics research group at the Department. The group is part of Helsinki Institute for Information Technology HIIT in Aalto University. The group is a member of the Finnish Centre of Excellence in Adaptive Informatics Research (AIRC) and the new Finnish Centre of Excellence in Computational Inference Research (COIN), and is also a member in the EU PASCAL2 network of excellence. The work will be supervised by the principal investigator of the project, academy research fellow Jaakko Peltonen. The work involves collaboration with Finnish researchers including members of the Statistical Machine Learning and Bioinformatics research group led by Prof. Samuel Kaski, and international researchers from the UK, Belgium, and the USA.

The research site is located on the Aalto University campus in Otaniemi, a short bus ride away from the centre of Finland’s capital Helsinki. Helsinki and the capital area are a great place to live, with numerous local attractions and events, scenic landscapes of small forests, islands and urban areas, a high standard of living, and excellent travel connections. Helsinki is the World Design Capital in 2012.

Department website: http://ics.tkk.fi/en/
HIIT website: http://www.hiit.fi/
Research group website: http://research.ics.tkk.fi/mi/
Principal investigator website: http://users.ics.tkk.fi/jtpelto/

Required and Desired Qualities of the Applicant:

A successful applicant must have a MSc degree in computer science, electrical engineering, mathematics, physics, or a related field. It is also possible to start as a research assistant working on one’s Master’s thesis.
– A strong mathematical background and an interest in probabilistic modeling and/or machine learning are necessary.
– An interest in some of the following topics is essential: dimensionality reduction, manifold learning, visualization, and multi-task learning. Experience in these topics is an advantage.
– A strong study record and strong track record in research are advantages.
– Good programming skills in languages such as C/C++/Matlab/R/Python and good written and spoken communication skills are desired.

Contract Details:

The salary will be determined based on the Aalto University salary system (2200-3200 euro per month before tax for a doctoral student depending on qualifications and performance). The initial appointment will be for one year. Extension will be possible depending on the availability of funding.

Application Procedure:

The application deadline is 31 October 2011, 3pm Finnish local time. The application materials must include a curriculum vitae, a copy of study records, contact details of at least two references, and any other materials deemed relevant. Applications must be submitted to the Registry of Aalto University, no later than the deadline. Applications must be submitted preferably by email to kirjaamo@aalto.fi or alternatively by physical mail to The Registry of Aalto University, Aalto University, P.O.Box 11000, FI-00076 Aalto, Finland (street address Otakaari 1, Espoo).

Candidates may be asked for an interview at Aalto University or via phone or Skype.

For additional information, please contact academy research fellow Jaakko Peltonen or HR Coordinator Stefan Ehrstedt. E-mail: firstname.lastname@aalto.fi
See also this ad online: http://www.aalto.fi/en/current/jobs/teaching_and_research/doctoral_student-research_assistant/

CFC: NIPS workshop on “Cosmology Meets Machine Learning”

Call for Contributions

NIPS 2011 Workshop on
“Cosmology Meets Machine Learning”
Sierra Nevada, Spain, December 16 or 17, 2011.

URL: http://webdav.is.mpg.de/pixel/cmml-nips2011.html

Submission for contributions is now open.
For more information please visit the meeting webpage.

Join us for an exciting program including invited talks by:

* Prof. Dr. Anthony Tyson, UC Davis
* Prof. Dr. Alexandre Refregier, ETH Zurich
* Prof. Dr. Jean-Luc Starck, CEA Saclay Paris
* Prof. Dr. David Hogg, New York University

Important Dates
—————

* November 2, 2011 Abstract submission deadline
* November 12, 2011 Notification of acceptance
* December 16 or 17, 2011 Workshop

Description
———–

Cosmology aims at the understanding of the universe and its evolution
through scientific observation and experiment and hence addresses one
of the most profound questions of human mankind. With the
establishment of robotic telescopes and wide sky surveys cosmology
already now faces the challenge of evaluating vast amount of data.

Multiple projects will image large fractions of the sky in the next
decade, for example the Dark Energy Survey will culminate in a
catalogue of 300 million objects extracted from peta-bytes of
observational data. The importance of automatic data evaluation and
analysis tools for the success of these surveys is undisputed.

Many problems in modern cosmological data analysis are tightly related
to fundamental problems in machine learning, such as classifying stars
and galaxies and cluster finding of dense galaxy populations. Other
typical problems include data reduction, probability density
estimation, how to deal with missing data and how to combine data from
different surveys.

An increasing part of modern cosmology aims at the development of new
statistical data analysis tools and the study of their behaviour and
systematics often not aware of recent developments in machine learning
and computational statistics.

Therefore, the objectives of this workshop are two-fold:

(i) The workshop aims to bring together experts from the Machine
Learning and Computational Statistics community with experts in the
field of cosmology to promote, discuss and explore the use of machine
learning techniques in data analysis problems in cosmology and to
advance the state of the art.

(ii) By presenting current approaches, their possible limitations, and
open data analysis problems in cosmology, this workshop aims to
encourage scientific exchange and to foster collaborations among the
workshop participants.

Submission Instructions
———————–

We invite submission of abstracts on topics in the following areas:

* challenging problems in cosmology data analysis
* applications of machine learning methods in cosmological data analysis problems

Submissions should not exceed 200 words and will be judged on
technical merit, the potential to generate discussion, and their
ability to foster collaboration within the workshop participants.
Accepted papers will be presented at the poster session with an
additional poster spotlight presentation. One author of every accepted
paper has to attend the workshop to present poster and spotlight talk.

Submissions should be sent to cmml.nips2011(at)gmail.com

Organizing Committee
——————–

Michael Hirsch, Universtiy College London
Sarah Bridle, University College London
Stefan Harmeling, Max Planck Institute for Intelligent Systems
Phil Marshall, Oxford University
Mark Girolami, University College London
Bernhard Schoelkopf, Max Planck Institute for Intelligent Systems

DEADLINE EXTENSION – EACL 2012: call for tutorial proposals

** Deadline extension: October 13th, 2011

Proposals are invited for the Tutorial Program of the 13th Conference
of the European Chapter of the Association for Computational
Linguistics (EACL 2012), to be held in Avignon, France, from April 23
to April 27, 2012. The selected tutorials will be given on the Monday
and Tuesday preceding the main conference (April 23 and 24).

EACL 2012 seeks proposals for tutorials in all areas of computational
linguistics, broadly conceived to include disciplines such as
linguistics (including phycholinguistics and other subfields), speech,
information retrieval and multimodal processing.

We particularly welcome (1) tutorials which cover advances in newly
emerging areas not previously covered in an (E)ACL related tutorial,
or (2) tutorials which provide introductions into related fields which
are potentially relevant for the CL community (e.g. bioinformatics,
social media, human language processing, machine learning
techniques). In order to gather a widespread audience, the interest of
the tutorial and the quality of the instructors will also be taken
into account.

REMUNERATION

Remuneration for tutorials is regulated by ACL policies:
http://aclweb.org/adminwiki/index.php?title=Policy_on_tutorial_teacher_payment

The conversion to euros will be done as follows: €550 for up to 20
people, plus €25 per person for registrants between 21 to 50, plus €18
per person for registrants greater than 50.

Please NOTE: Remuneration for Tutorial presenters is fixed according
to the above policy and does not cover registration fees for the
main conference.

SUBMISSION DETAILS

Proposals for tutorials should contain:

1. A title and brief description of the tutorial content and its
relevance to the ACL community (not more than 2 pages).

2. A brief outline of the tutorial structure showing that the
tutorial’s core content can be covered in a three-hour slot
(excluding a coffee break). In exceptional cases six-hour tutorial
slots are available as well.

3. The names, postal addresses, phone numbers, and email addresses of
the tutorial instructors, including a one-paragraph statement of
their research interests and areas of expertise.

4. A list of previous venues and approximate audience sizes, if the
same or a similar tutorial has been given elsewhere; otherwise an
estimate of the audience size.

5. A description of special requirements for technical equipment
(e.g. internet access). Proposals should be submitted by
electronic mail, in plain ASCII text, to tutorials at eacl2012 dot
org, no later than September 30th 2011.

The subject line should be: “EACL 2012 TUTORIAL PROPOSAL”.

PLEASE NOTE:

– only proposals submitted by e-mail will be taken into account.

– you will receive email confirmation from us that your proposal has
been received. If you do not receive this confirmation 24 hours
after sending the proposal, please contact us personally using both
e.agirre at ehu dot es and lieve.macken at hogent dot be

TUTORIAL SPEAKER RESPONSIBILITIES

Accepted tutorial speakers will be notified by November 3rd, 2011,
and must then provide abstracts of their tutorials for inclusion in
the conference registration material by December 16th, 2011. The
description should be in two formats: an ASCII version that can be
included in email announcements and published on the conference web
site, and a PDF version for inclusion in the electronic proceedings
(detailed instructions to follow).

Tutorial speakers must provide tutorial materials, at least containing
copies of the course slides as well as a bibliography for the material
covered in the tutorial, by February 1st, 2012.

IMPORTANT DATES

Submission deadline for tutorial proposals: October 13th, 2011
Notification of acceptance: November 3rd, 2011
Tutorial descriptions due: December 16th, 2011
Tutorial course material due: February 1st, 2012
Tutorial dates: April 23-24, 2012

TUTORIAL CHAIRS

Eneko Agirre, University of the Basque Country, Spain
Lieve Macken, University College Ghent, Belgium

Please send inquiries concerning EACL 2012 tutorials to:
tutorials at eacl2012 dot org