I did some work some years ago on this topic, and I can recommend a
couple of books. One can go much further with some general solid
knowledge of machine learning in itself.
The book by Hastie, Tibshirani, and Friedman, The Elements of
Statistical Learning, now at its 2nd edition, changed my thinking
completely. This book treats machine learning algorithms from the
perspective of applied statistics (mostly through an aggressive use of
regression methodology), showing how applied statistical thinking can
inform this apparently different area. Concise, crisp, and
intellectually satisfying.
Probably the most successful product of machine learning is the
support vector machine. A very readable and useful compact book is An
Introduction to Support Vector Machines by Cristianini and
Shawe-Taylor, with an update in 2004. The fascinating story about SVM
is that it developed behind the iron curtain in isolation through the
interbreeding of the Russian probabilist school and artificial
intelligence. One can read about this in the book by its creator,
Vladimir Vapnik, The Nature of Statistical Learning Theory. After the
wall came down, Vapnik moved to AT&T, then to NEC.
Best,
Roberto Viviani
University of Ulm, Germany
> I don't know of any courses, but the Video Lectures web site has some
> nice stuff. Maybe have a look through:
>
> http://videolectures.net/fmri06_whistler/
> http://videolectures.net/Top/Computer_Science/Machine_Learning/
>
> Best regards,
> -John
>
> On Mon, 2010-01-18 at 15:53 +0100, Wouter De Baene wrote:
>> Dear all,
>>
>> Although my question is not directly SPM-related, it is related to this
>> previous mail. Does anybody know of short courses on pattern recognition
>> related to fMRI?
>>
>> Best regards,
>> Wouter De Baene
>>
>>
>> On 18/01/10 15:40, "Jonas Richiardi" <[log in to unmask]> wrote:
>>
>> > Dear Colleagues, please accept our apologies for multiple postings.
>> >
>> > * CALL FOR PAPERS *
>> >
>> > 1st ICPR Workshop on Brain Decoding:
>> > Pattern Recognition Challenges in Neuroimaging
>> >
>> > In conjunction with the
>> > 20th International Conference on Pattern Recognition
>> > August 22, 2010, Istanbul, Turkey
>> >
>> > http://miplab.epfl.ch/icpr2010/
>> >
>> > Many of the challenges facing brain decoding are also highly relevant to
>> > the field of pattern recognition at large - classification of multivariate
>> > time-series, dimensionality reduction, and causal modelling to name a few.
>> >
>> > This workshop wants to bring together researchers in pattern recognition
>> > and neuroimaging for fruitful exchanges of experiences and recent
>> > developments in brain decoding, both on the methodological side and the
>> > application side. The temporal and spatial proximity with the leading
>> > pattern recognition conference ICPR is a great opportunity to learn about
>> > newest scientific developments in order to expand brain decoding
>> > methodology.
>> >
>> > SCOPE
>> >
>> > The topics of interest include:
>> >
>> > * Data representations for brain decoding (MRI/fMRI/EEG/...)
>> > - Voxel and feature selection
>> > - Linear and non-linear dimensionality reduction
>> > - Sparse timecourse representations
>> >
>> > * Classifiers for high-dimensional learning
>> > - Regularisation schemes
>> > - Subspace methods
>> > - Interpretability and validation
>> >
>> > * Applications of brain decoding
>> > - Visual processing
>> > - Man-machine interfaces
>> > - Clinical applications
>> > - Deception detection
>> >
>> > SUBMISSION AND PROCEEDINGS
>> >
>> > Authors should prepare full 4-pages papers (double-column, IEEE style).
>> > Manuscripts will be evaluated by 2 reviewers.
>> >
>> > Proceedings will be published by IEEE Computer Science Society in
>> > electronic format. They will be permantently available on the IEEExplore
>> > and IEEE CS Digital Library online repositories, and indexed in IEE
>> > INSPEC, EI Compendex (Elsevier), and others.
>> >
>> > DATES AND DEADLINES
>> >
>> > - April 1, 2010: Paper submission deadline
>> > - May 1, 2010: Acceptance Notification
>> > - May 14, 2010: Early registration deadline
>> > - June 1, 2010: Camera-ready paper due
>> > - August 22, 2010: Workshop
>> >
>> > PROGRAMME COMMITTEE
>> >
>> > Rafeef Abugharbieh, University of British Columbia (Canada)
>> > Tulay Adali, University of Maryland, Baltimore County (USA)
>> > John Ashburner, University College London (UK)
>> > Patricia Besson, CNRS (France)
>> > Mick Brammer, King¹s College London (UK)
>> > Vince Calhoun, Yale University (USA)
>> > Thomas Ethofer, University of Tübingen (Germany)
>> > Christian Gaser, University of Jena (Germany)
>> > Ghassan Hamarneh, Simon Fraser University (Canada)
>> > David Hardoon, Institute for Infocomm Research (Singapore)
>> > Krzysztof Kryszczuk, IBM Research Zürich (Switzerland)
>> > Raghu Machiraju, Ohio State University (USA)
>> > Andre Marquand, King¹s College London (UK)
>> > Julien Meynet, Yahoo! research (France)
>> > Torsten Möller, Simon Fraser University (Canada)
>> > Joao Sato, Federal University of ABC (Brazil)
>> > Sophie Schwartz, University of Geneva (Switzerland)
>> > Stephen Strother, University of Toronto (Canada)
>> > Bertrand Thirion, NeuroSpin Paris (France)
>> > Marc Van Hulle, K.U. Leuven (Belgium)
>> > Patrik Vuilleumier, University of Geneva (Switzerland)
>> >
>> > ORGANISING COMMITTEE
>> >
>> > Jonas Richiardi, Ecole Polytechnique Fédérale de Lausanne (EPFL)
>> > and University of Geneva (Swizerland)
>> > Dimitri Van De Ville, Ecole Polytechnique Fédérale de Lausanne (EPFL)
>> > and University of Geneva (Switzerland)
>> > Christos Davatzikos, University of Pennsylvania (USA)
>> > Janaina Mourão-Miranda, University College London
>> > and King's College London (UK)
>> >
>> > details: http://miplab.epfl.ch/icpr2010/
>> > contact: wbd2010 _AT_ listes.epfl.ch
>>
>
>
> --
> John Ashburner <[log in to unmask]>
>
|