Tuesday, October 20, 2015 - 11:30am - 12:20pm
Shie Mannor (Technion-Israel Institute of Technology)
We consider the problem of detecting epidemics in graphs when the indication if a node is infected
is extremely noisy. We show that with even overwhelming noise the structure of the network can still
be used to detect epidemics. Our approach uses local algorithms for detection and only a fairly
loose information about the network structure is needed. Our analysis relies on percolation theory and
tools from analysis of extreme events of diffusion processes over graphs.
Tuesday, September 24, 2013 - 9:00am - 9:50am
Rebecca Willett Lu (University of Wisconsin, Madison)
In this talk, I will describe a novel approach to change-point
detection when the observed high-dimensional data may have missing
elements. The performance of classical methods for change-point
detection typically scales poorly with the dimensionality of the data,
so that a large number of observations are collected after the true
change-point before it can be reliably detected. Furthermore, missing
components in the observed data handicap conventional approaches. The
Subscribe to RSS - Detection