Abstract | ||
---|---|---|
This paper addresses the problem of learning from very large databases where batch learning is impractical or even infeasible. Bootstrap is a popular tech- nique applicable in such situations. We show that sampling strategy used for bootstrapping has a significant impact on the resulting classifier perfor- mance. We design a new general sampling strategy "quasi-random weighted sampling + trimming" (QWS+) that includes well established strategies as special cases. The QWS+ approach minimizes the variance of hypothesis er- ror estimate and leads to significant improvement in performance compared to standard sampling techniques. The superior performance is demonstrated on several problems including profile and frontal face detection. |
Year | Venue | Keywords |
---|---|---|
2008 | BMVC | sampling technique,very large database,face detection |
Field | DocType | Citations |
Pattern recognition,Bootstrapping,Computer science,Artificial intelligence,Boosting (machine learning),Sampling (statistics),Face detection,Classifier (linguistics),Trimming,Machine learning,Bootstrapping (electronics) | Conference | 40 |
PageRank | References | Authors |
2.70 | 15 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Zdenek Kalal | 1 | 1023 | 36.85 |
Jiri Matas | 2 | 4313 | 234.68 |
Krystian Mikolajczyk | 3 | 7280 | 625.08 |