BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Event-based representations for Electromagnetic Brain Signals
DTSTART:20250707T093500Z
DTEND:20250707T100500Z
DTSTAMP:20260612T074900Z
UID:indico-event-14492@indico.math.cnrs.fr
DESCRIPTION:Speakers: Thomas Moreau\n\nThe quantitative analysis of non-in
 vasive electrophysiology signals from electroencephalography (EEG) and mag
 netoencephalography (MEG) often boils down to the identification of certai
 n types of events and their distribution in the signal. The events are cha
 racterized by their temporal patterns\, such as evoked responses\, transie
 nt bursts of neural oscillations\, but also blinks or heartbeats for data 
 cleaning. Given these events and patterns\, a natural question is to estim
 ate how their occurrences are modulated by certain cognitive tasks and exp
 erimental manipulations. In this talk\, I will present contributions to th
 e analysis of event-related neural responses using point-process models. W
 hile PP has been used in neuroscience in the past\, in particular for sing
 le cell recordings (spike trains)\, techniques such as CDL make them amena
 ble to human studies based on EEG/MEG signals. A particular focus will be 
 on the development of methods that scale to the large dimensionality of th
 e data in the neuroscience context\, with efficient optimization procedure
 s and robustness to noisy observation\, with early results on M/EEG datase
 ts. \n\nhttps://indico.math.cnrs.fr/event/14492/
URL:https://indico.math.cnrs.fr/event/14492/
END:VEVENT
END:VCALENDAR