Mathematics and algorithmics of process detection

George Cybenko
Dartmouth University
School of Engineering

Multiple process detection involves identifying instances of several dynamical processes and estimating their states from a sequence of unlabeled, noisy and ambiguous observations of the processes. Multiple process detection arises in many applications areas of current interest, including infrastructure management and security, sensor networks, autonomic systems and attention focusing in cognitive systems. This talk will provide an overview of the theory and applications of process detection problems with
Illustrative examples.

A software implementation of a general-purpose process detection system will be presented as well.


Presentation (PowerPoint File)
Video of Talk (RealPlayer File)

Back to Graduate Summer School: Intelligent Extraction of Information from Graphs and High Dimensional Data