Abstract | ||
---|---|---|
The performance of active learning (AL) is crucially influenced by the existence of outliers in input samples. In this paper, we propose a robust pool-based AL measure based on the density power divergence. It is known that the density power divergence can be accurately estimated even under the existence of outliers within data. We further derive an AL scheme based on an asymptotic statistical analysis on the M-estimator. The performance of the proposed framework is investigated empirically using artificial and real-world data. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1007/978-3-642-34487-9_72 | ICONIP (3) |
Keywords | Field | DocType |
real-world data,robust pool-based al measure,proposed framework,active learning,linear regression,al scheme,density power divergence,asymptotic statistical analysis,robust active learning,input sample,regression | Divergence,Active learning,Pattern recognition,Regression,Outlier,Artificial intelligence,Statistics,Machine learning,Mathematics,Statistical analysis,Linear regression | Conference |
Volume | ISSN | Citations |
7665 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 4 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yasuhiro Sogawa | 1 | 75 | 4.85 |
Tsuyoshi Ueno | 2 | 14 | 4.37 |
Kawahara, Yoshinobu | 3 | 317 | 31.30 |
Takashi Washio | 4 | 1775 | 190.58 |