BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Machine Learning for Neural Engineering
DTSTART:20160201T140000
DTEND:20160201T150000
DTSTAMP:20260406T150536Z
UID:b6058623506f4553df31ff97d712c59cb5403fb2c0c17e5f9c0272e3
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr. Moritz Grosse-Wentrup (Max Planck Institute)\nScience trad
 itionally progresses by developing mechanistic models that capture cause-e
 ffect relations. In cognitive neuroscience\, such models of are not yet av
 ailable. While this is often seen as a major shortcoming\, I will argue th
 at we do not actually need mechanistic models to translate research in cog
 nitive neuroscience into clinical applications. I will present machine lea
 rning methods that can infer causal hypotheses directly from empirical dat
 a. I will demonstrate the utility of this novel conceptual approach on two
  applications. First\, I will show how it has led to a new class of brain-
 computer interfaces (BCIs) for communication with patients in late stages
  of amyotrophic lateral sclerosis (ALS). Second\, I will present a novel a
 pproach to BCI-based motor rehabilitation that considers the global config
 uration of brain rhythms and its relation to motor deficits.\nA live trans
 mission will be available on the EPFL campus in SV3715
LOCATION:CB H8-144.165 + SV3715 http://cnp.epfl.ch/files/content/sites/cnp
 /files/Seminars/PlanSallesCampusBiotechH8-01.pdf
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR