BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//hacksw/handcal//NONSGML v1.0//EN
BEGIN:VEVENT
UID:node-3745@cbmm.mit.edu
DTSTAMP:20181012T200000Z
DTSTART:20181012T200000Z
DTEND:20181012T210000Z
SUMMARY:Brains, Minds + Machines Seminar Series: Modal-Set Estimation using kNN graphs, and Applications to Clustering
DESCRIPTION:Speaker: Prof. Samory Kpotufe, Princeton University\nEvent Description: Abstract: Estimating the mode or modal-sets (i.e. \nextrema points or surfaces) of an unknown density from sample is a basic \nproblem in data analysis. Such estimation is relevant to other problems such \nas clustering, outlier detection, or can simply serve to identify \nlow-dimensional structures in high dimensional-data (e.g. point-cloud data \nfrom medical-imaging, astronomy, etc). Theoretical work on mode-estimation \nhas largely concentrated on understanding its statistical difficulty, while \nless attention has been given to implementable procedures. Thus, theoretical \nestimators, which are often statistically optimal, are for the most part hard \nto implement. Furthermore for more general modal-sets (general extrema of any \ndimension and shape) much less is known, although various existing procedures \n(e.g. for manifold-denoising or density-ridge estimation) have similar \npractical aim. I’ll present two related contributions of indep...
LOCATION:Singleton Auditorium (MIT 46-3002) 43 Vassar Street, Cambridge MA 02139\n
END:VEVENT
END:VCALENDAR