A parametric procedure for learning with an imperfect teacher (Corresp.)
- 1 March 1972
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Theory
- Vol. 18 (2) , 300-302
- https://doi.org/10.1109/tit.1972.1054780
Abstract
Pattern recognition problems involving learning with a bad teacher or learning without a teacher require the updating of the conditional densities of unknown parameters using a mixture of probability density functions. Mixtures of density functions in general are not reproducing and hence the computations are infeasible. For learning without a teacher, a computationally feasible scheme has been suggested by Agrawala [1]. The learning procedure proposed by Agrawala makes use of a probabilistic labeling scheme. The probabilistic labeling scheme is extended to allow the use of reproducing densities for a large class of problems, including the problem of learning with an imperfect teacher.Keywords
This publication has 4 references indexed in Scilit:
- Learning with a probabilistic teacherIEEE Transactions on Information Theory, 1970
- Learning to recognize patterns without a teacherIEEE Transactions on Information Theory, 1967
- A note on the iterative application of Bayes' ruleIEEE Transactions on Information Theory, 1965
- Adaptive communication receiversIEEE Transactions on Information Theory, 1965