Adaptive online learning of optimal decision boundary using active sampling

Author

Park, Jong-Min ; Hu, Yu Hen

Author_Institution

Dept. of Electr. & Comput. Eng., Wisconsin Univ., Madison, WI, USA

fYear

1996

fDate

4-6 Sep 1996

Firstpage

253

Lastpage

262

Abstract

An adaptive online learning method is presented to facilitate pattern classification using active sampling to identify optimal decision boundary for a stochastic oracle with a minimum number of training samples. The strategy of sampling at the current estimate of the decision boundary is shown to be optimal in the sense that the probability of convergence toward the true decision boundary at each step is maximized, offering theoretical justification on the popular strategy of category boundary sampling used by many query learning algorithms

Keywords

adaptive systems; convergence of numerical methods; estimation theory; iterative methods; learning (artificial intelligence); optimisation; pattern classification; perceptrons; probability; real-time systems; stochastic processes; active learning; active sampling; adaptive online learning; category boundary sampling; convergence; optimal decision boundary; optimisation; pattern classification; perceptrons; probability; query learning algorithms; stochastic oracle; Convergence; Cost function; Design for experiments; Drives; Learning systems; Pattern classification; Pattern recognition; Probability; Sampling methods; Stochastic processes;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks for Signal Processing [1996] VI. Proceedings of the 1996 IEEE Signal Processing Society Workshop

Conference_Location

Kyoto

ISSN

1089-3555

Print_ISBN

0-7803-3550-3

Type

conf

DOI

10.1109/NNSP.1996.548355

Filename

548355