NIPS 2011 - NIPS 2011 Workshop on Machine Learning and Inference in Neuroimaging
View: 5076
Website |
Edit Freely
Category NIPS 2011
Deadline: September 30, 2011 | Date: December 16, 2011-December 17, 2011
Venue/Country: Granada, Spain
Updated: 2011-09-02 07:06:51 (GMT+9)
Call For Papers - CFP
Modern multivariate statistical methods have been increasingly appliedto various problems in neuroimaging, including “mind reading”, “brainmapping”, clinical diagnosis and prognosis. Multivariate patternanalysis (MVPA) is a promising machine-learning approach fordiscovering complex relationships between high-dimensional signals(e.g., brain images) and variables of interest (e.g., external stimuliand/or brain's cognitive states). Modern multivariate regularizationapproaches can overcome the curse of dimensionality and produce highlypredictive models even in high-dimensional, low-sample scenariostypical in neuroimaging (e.g., 10 to 100 thousands of voxels and justa few hundreds of samples).However, despite the rapidly growing number of neuroimagingapplications in machine learning, its impact on how theories of brainfunction are construed has received little consideration. Accordingly,machine-learning techniques are frequently met with skepticism in thedomain of cognitive neuroscience. In this workshop, we intend toinvestigate the implications that follow from adopting machine-learning methods for studying brain function. In particular, thisconcerns the question how these methods may be used to representcognitive states, and what ramifications this has for consequenttheories of cognition. Besides providing a rationale for the use ofmachine-learning methods in studying brain function, a further goal ofthis workshop is to identify shortcomings of state-of-the-artapproaches and initiate research efforts that increase the impact ofmachine learning on cognitive neuroscience.Moreover, from the machine learning perspective, neuroimaging is arich source of challenging problems that can facilitate development ofnovel approaches. For example, feature extraction and featureselection approaches become particularly important in neuroimaging,since the primary objective is to gain a scientific insight ratherthan simply learn a ``black-box'' predictor. However, unlike someother applications where the set features might be quite well-exploredand established by now, neuroimaging is a domain where a machine-learning researcher cannot simply "ask a domain expert what featuresshould be used", since this is essentially the question the domainexpert themselves are trying to figure out. While the currentneuroscientific knowledge can guide the definition of specialized'brain areas', more complex patterns of brain activity, such as spatio-temporal patterns, functional network patterns, and other multivariatedependencies remain to be discovered mainly via statistical analysis.The list of open questions of interest to the workshop includes, butis not limited to the following:● How can we interpret results of multivariate models in aneuroscientific context?● How suitable are MVPA and inference methods for brain mapping?● How can we assess the specificity and sensitivity?● What is the role of decoding vs. embedded or separate featureselection?● How can we use these approaches for a flexible and usefulrepresentation of neuroimaging data?● What can we accomplish with generative vs. discriminative modelling?Workshop Format:In this two-day workshop we will explore perspectives and novelmethodology at the interface of Machine Learning, Inference,Neuroimaging and Neuroscience. We aim to bring researchers frommachine learning and neuroscience community together, in order todiscuss open questions, identify the core points for a number of thecontroversial issues, and eventually propose approaches to solvingthose issues.The workshop will be structured around 3 main topics:- machine learning and pattern recognition methodology- causal inference in neuroimaging- linking machine learning, neuroimaging and neuroscienceEach session will be opened by 2-3 invited talks, and an in depthdiscussion. This will be followed by original contributions. Originalcontributions will also be presented and discussed during a postersession. The workshop will end with a panel discussion, during whichwe will address specific questions, and invited speakers will openeach segment with a brief presentation of their opinion.This workshop proposal is part of the PASCAL2 Thematic Programme onCognitive Inference and Neuroimaging (http://mlin.kyb.tuebingen.mpg.de/).Paper Submission:We seek for submission of original research papers. The length of thesubmitted papers should not exceed 4 pages in Springer format (hereare the LaTeX2e style files). We aim at publishing accepted paperafter the workshop in a proceedings volume that contains full papers,together with review papers by the invited speakers. Authors areexpected to prepare a full 8 page paper for the final camera readyversion after the workshop.Important dates:- September 30, 2011 - paper submission- October 15th, 2011 - notification of acceptance/rejection- December 16th - 17th - Workshop in Sierra Nevada, Spain, followingthe NIPS conferenceInvited Speakers:Polina Golland (MIT, US)James V. Haxby (Dartmouth College, US)Tom Mitchell (CMU, US)Daniel Rueckert (Imperial College, UK)Peter Spirtes (CMU, US)Gaël Varoquaux (Neurospin/INRIA, France)Program Committee:Guillermo Cecchi (IBM T.J. Watson Research Center)Melissa Carroll (Google)Moritz Grosse-Wentrup (Max Planck Institute for Intelligent Systems,Tübingen, Germany)*James V. Haxby (Dartmouth College, USA, University of Trento, Italy)Georg Langs (Medical University of Vienna)*Bjoern Menze (ETH Zuerich, CSAIL, MIT)Janaina Mourao-Miranda (University College London, United Kingdom)Vittorio Murino (University of Verona/Istituto Italiano di Tecnologia,Italy)Francisco Pereira (Princeton University)Irina Rish (IBM T.J. Watson Research Center)*Mert Sabuncu (Harvard Medical School)Bertrand Thirion (INRIA, NEUROSPIN)
Keywords: Accepted papers list. Acceptance Rate. EI Compendex. Engineering Index. ISTP index. ISI index. Impact Factor.
Disclaimer: ourGlocal is an open academical resource system, which anyone can edit or update. Usually, journal information updated by us, journal managers or others. So the information is old or wrong now. Specially, impact factor is changing every year. Even it was correct when updated, it may have been changed now. So please go to Thomson Reuters to confirm latest value about Journal impact factor.