Delannay, Nicolas
[UCL]
Archambeau, Cédric
[UCL]
Verleysen, Michel
[UCL]
K-nearest neighbors relies on the definition of a global metric. In contrast, discriminant adaptive nearest neighbor (DANN) computes a different metric at each query point based on a local linear discriminant analysis. In this paper, we propose a technique to automatically adjust the hyper-parameters in DANN by the optimization of two quality criteria. The first one measures the quality of discrimination, while the second one maximizes the local class homogeneity. We use a Bayesian formulation to prevent over-fitting.
Bibliographic reference |
Delannay, Nicolas ; Archambeau, Cédric ; Verleysen, Michel. Automatic adjustment of discriminant adaptive nearest neighbor.18th International Conference on Pattern Recognition (ICPR 2006) (Hong Kong (China), du 20/08/2006 au 24/08/2006). In: Tang, Y.Y.; Wang, S.P.; Lorette, G.L.; Yeung, D.S.; Yan, H.;, Proceedings of the 18th International Conference on Pattern Recognition (ICPR 2006), IEEE comput. soc2006, p.4 pages |
Permanent URL |
http://hdl.handle.net/2078.1/67916 |