Online Learning of Non-stationary Sequences
dc.contributor.author | Monteleoni, Claire | |
dc.contributor.author | Jaakkola, Tommi | |
dc.date.accessioned | 2005-12-22T02:40:44Z | |
dc.date.available | 2005-12-22T02:40:44Z | |
dc.date.issued | 2005-11-17 | |
dc.identifier.other | MIT-CSAIL-TR-2005-074 | |
dc.identifier.other | AIM-2005-032 | |
dc.identifier.uri | http://hdl.handle.net/1721.1/30584 | |
dc.description.abstract | We consider an online learning scenario in which the learner can make predictions on the basis of a fixed set of experts. We derive upper and lower relative loss bounds for a class of universal learning algorithms involving a switching dynamics over the choice of the experts. On the basis of the performance bounds we provide the optimal a priori discretization of the switching-rate parameter that governs the switching dynamics. We demonstrate the algorithm in the context of wireless networks. | |
dc.format.extent | 8 p. | |
dc.format.extent | 10189026 bytes | |
dc.format.extent | 760649 bytes | |
dc.format.mimetype | application/postscript | |
dc.format.mimetype | application/pdf | |
dc.language.iso | en_US | |
dc.relation.ispartofseries | Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory | |
dc.subject | AI | |
dc.subject | online learning | |
dc.subject | regret bounds | |
dc.subject | non-stationarity | |
dc.subject | HMM | |
dc.subject | wireless networks | |
dc.title | Online Learning of Non-stationary Sequences |