June 2011 - Lisa_seminaires - mailman.iro.umontreal.ca

Reminder: CIFAR machine learning seminar today @ 14h30, AA3195
by Guillaume Desjardins 23 Jun '11

23 Jun '11

---------- Forwarded message ---------- From: Guillaume Desjardins <guillaume.desjardins(a)gmail.com> Date: Mon, Jun 20, 2011 at 3:56 PM Subject: Update: CIFAR machine learning seminar Thurs. June 23rd @ 14h30, AA3195 To: lisa_seminaires(a)iro.umontreal.ca Aapo Hyvarinen's talk, scheduled for Thurs. June 23rd @ 14h30, will take place in AA3195 (Pavillon Andre-Aisenstadt, room 3195, University of Montreal). On Fri, Jun 17, 2011 at 6:07 PM, Guillaume Desjardins <guillaume.desjardins(a)gmail.com> wrote: > I am delighted to announce the following talk by Aapo Hyvarinen, which > will take place on Thursday June 23rd at 14h30. The talk will be held > at the University of Montreal, with the exact room to be announced at > a later date. Hope to see you there ! > > Title: Unsupervised machine learning for analysis of EEG and MEG at rest > Speaker: Aapo Hyvarinen > > Abstract: Recently, a lot of brain imaging has concentrated on > analyzing brain activity at rest, i.e. when the subject is not doing > anything particular and does not receive specific stimulation. When > using functional magnetic resonance imaging (fMRI), such analysis is > typically done by independent component analysis (ICA). However, there > has not been very much work on analysing resting activity measured by > EEG or MEG. We have recently developed various methods to analyse > EEG/MEG data at rest. First, we have created new variants of ICA to > more completely exploit the statistical structure of EEG/MEG data. > Second, we have developed tests of the statistical significance of the > independent components. Third, we have a framework for analysis of > causality (connectivity) which uses the non-Gaussianity of the data > and hopefully it can be used to analyse the connectivity of the > independent components. >

1 0

Update: CIFAR machine learning seminar Thurs. June 23rd @ 14h30, AA3195
by Guillaume Desjardins 20 Jun '11

20 Jun '11

Aapo Hyvarinen's talk, scheduled for Thurs. June 23rd @ 14h30, will take place in AA3195 (Pavillon Andre-Aisenstadt, room 3195, University of Montreal). On Fri, Jun 17, 2011 at 6:07 PM, Guillaume Desjardins <guillaume.desjardins(a)gmail.com> wrote: > I am delighted to announce the following talk by Aapo Hyvarinen, which > will take place on Thursday June 23rd at 14h30. The talk will be held > at the University of Montreal, with the exact room to be announced at > a later date. Hope to see you there ! > > Title: Unsupervised machine learning for analysis of EEG and MEG at rest > Speaker: Aapo Hyvarinen > > Abstract: Recently, a lot of brain imaging has concentrated on > analyzing brain activity at rest, i.e. when the subject is not doing > anything particular and does not receive specific stimulation. When > using functional magnetic resonance imaging (fMRI), such analysis is > typically done by independent component analysis (ICA). However, there > has not been very much work on analysing resting activity measured by > EEG or MEG. We have recently developed various methods to analyse > EEG/MEG data at rest. First, we have created new variants of ICA to > more completely exploit the statistical structure of EEG/MEG data. > Second, we have developed tests of the statistical significance of the > independent components. Third, we have a framework for analysis of > causality (connectivity) which uses the non-Gaussianity of the data > and hopefully it can be used to analyse the connectivity of the > independent components. >

1 0

CIFAR machine learning seminar Thurs. June 23rd @ 14h30, room TBA
by Guillaume Desjardins 17 Jun '11

17 Jun '11

I am delighted to announce the following talk by Aapo Hyvarinen, which will take place on Thursday June 23rd at 14h30. The talk will be held at the University of Montreal, with the exact room to be announced at a later date. Hope to see you there ! Title: Unsupervised machine learning for analysis of EEG and MEG at rest Speaker: Aapo Hyvarinen Abstract: Recently, a lot of brain imaging has concentrated on analyzing brain activity at rest, i.e. when the subject is not doing anything particular and does not receive specific stimulation. When using functional magnetic resonance imaging (fMRI), such analysis is typically done by independent component analysis (ICA). However, there has not been very much work on analysing resting activity measured by EEG or MEG. We have recently developed various methods to analyse EEG/MEG data at rest. First, we have created new variants of ICA to more completely exploit the statistical structure of EEG/MEG data. Second, we have developed tests of the statistical significance of the independent components. Third, we have a framework for analysis of causality (connectivity) which uses the non-Gaussianity of the data and hopefully it can be used to analyse the connectivity of the independent components.

1 0

Fwd: Séminaire RALI-OLST MITACS du 22 juin 2011
by Yoshua Bengio 16 Jun '11

16 Jun '11

Begin forwarded message: > From: Webmestre RALI <critter(a)iro.umontreal.ca> > Date: June 16, 2011 9:09:52 AM EDT (CA) > To: rali-olst(a)iro.umontreal.ca > Subject: Séminaire RALI-OLST MITACS du 22 juin 2011 > > > Annonce de séminaire RALI-OLST MITACS > > Envisioning Dynamic Quantum Clustering in Information Retrieval > Giorgio Maria Di Nunzio > University of Padua > > Le 22 juin 2011 à 11h30 > Salle 3195, Pavillon André-Aisenstadt > > ---- > Clustering is the problem of automatically organizing a large data > collection by partitioning the data into clusters, so the objects in > the > same cluster are more similar to one another than to objects in other > clusters. Textual document clustering aims at organizing the > documents of > a collection in a way that results in better search without > introducing > much extra cost and complexity. For instance, clustering strategies > have > been investigated in Information Retrieval to support alternative > organizations of the search results. In this talk I'll present a > study of > possible application of a recently proposed clustering method, named > Dynamic Quantum Clustering (DQC), to the problem of document > clustering. > This method works by analogy by considering data samples as > particles that > obey physics laws. In our case particles are the documents of the > collection. Some initial results will be presented. These results > analyze > the capability of the DQC to select highly relevant documents > relative to > a query of a user. > ---- > > > Consulter l'annonce du séminaire sur le Web : > http://rali.iro.umontreal.ca/sem/display_sem.php?semid=3432

1 0