Sequential anomaly detection in the presence of noise and limited feedback

dc.contributor.author	Raginsky, Maxim
dc.contributor.author	Willett, Rebecca
dc.date.accessioned	2009-08-13T15:00:44Z
dc.date.available	2009-08-13T15:00:44Z
dc.date.issued	2009-08-13T15:00:44Z
dc.identifier.uri	http://hdl.handle.net/10161/1286
dc.description.abstract	Recent work has established the efficacy of using online convex programming methods on exponential families in the context of sequential probability assignment. This paper describes methods which build upon that framework to handle noisy observations. Furthermore, the problem of detecting anomalous (i.e. rare) events by using the sequential probability assignments and limited feedback is presented.	en_US
dc.format.extent	264226 bytes
dc.format.mimetype	application/pdf
dc.language.iso	en_US	en_US
dc.relation.ispartofseries	ECE-2009-01	en_US
dc.subject	Filtering	en_US
dc.subject	universal prediction	en_US
dc.subject	individual sequences	en_US
dc.subject	anomaly detection	en_US
dc.title	Sequential anomaly detection in the presence of noise and limited feedback	en_US
dc.type	Technical Report	en_US
dc.department	Engineering