Online Learning of Non-stationary Sequences
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 p...
| Main Authors: | , |
|---|---|
| Language: | en_US |
| Published: |
2005
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/1721.1/30584 |
| _version_ | 1811095927210901504 |
|---|---|
| author | Monteleoni, Claire Jaakkola, Tommi |
| author_facet | Monteleoni, Claire Jaakkola, Tommi |
| author_sort | Monteleoni, Claire |
| collection | MIT |
| description | 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. |
| first_indexed | 2024-09-23T16:34:26Z |
| id | mit-1721.1/30584 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T16:34:26Z |
| publishDate | 2005 |
| record_format | dspace |
| spelling | mit-1721.1/305842019-04-11T06:23:37Z Online Learning of Non-stationary Sequences Monteleoni, Claire Jaakkola, Tommi AI online learning regret bounds non-stationarity HMM wireless networks 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. 2005-12-22T02:40:44Z 2005-12-22T02:40:44Z 2005-11-17 MIT-CSAIL-TR-2005-074 AIM-2005-032 http://hdl.handle.net/1721.1/30584 en_US Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory 8 p. 10189026 bytes 760649 bytes application/postscript application/pdf application/postscript application/pdf |
| spellingShingle | AI online learning regret bounds non-stationarity HMM wireless networks Monteleoni, Claire Jaakkola, Tommi Online Learning of Non-stationary Sequences |
| title | Online Learning of Non-stationary Sequences |
| title_full | Online Learning of Non-stationary Sequences |
| title_fullStr | Online Learning of Non-stationary Sequences |
| title_full_unstemmed | Online Learning of Non-stationary Sequences |
| title_short | Online Learning of Non-stationary Sequences |
| title_sort | online learning of non stationary sequences |
| topic | AI online learning regret bounds non-stationarity HMM wireless networks |
| url | http://hdl.handle.net/1721.1/30584 |
| work_keys_str_mv | AT monteleoniclaire onlinelearningofnonstationarysequences AT jaakkolatommi onlinelearningofnonstationarysequences |