Title :
On-line learning by active sampling using orthogonal decision support vectors
Author_Institution :
Dept. of Electr. & Comput. Eng., San Diego State Univ., CA, USA
Abstract :
Active-sampling-at-the-boundary method is applied using orthogonal decision support vectors to facilitate pattern classification in identifying optimal decision boundary for a stochastic oracle. The result of the active sampling near the boundary using these vectors is shown in comparison with active learning using random selection in the multi-dimensional decision hyperplane. This shows the optimality of boundary active sampling using decision support vectors in the case of non-separable linear decision hyperplanes in multi-dimensional space
Keywords :
learning (artificial intelligence); pattern classification; stochastic processes; active learning; active sampling; linear decision hyperplanes; multidimensional decision hyperplane; online learning; optimal decision boundary; orthogonal decision support vectors; pattern classification; random selection; stochastic oracle; Design for experiments; Learning systems; Machine learning; Machine learning algorithms; Monte Carlo methods; Pattern classification; Sampling methods; Stochastic processes; Training data; Vectors;
Conference_Titel :
Neural Networks for Signal Processing X, 2000. Proceedings of the 2000 IEEE Signal Processing Society Workshop
Conference_Location :
Sydney, NSW
Print_ISBN :
0-7803-6278-0
DOI :
10.1109/NNSP.2000.889410
