Optimal filters for attribute generation and machine learning
暂无分享,去创建一个
Extensions to inductive inference methods of machine learning are proposed which allow inference from dynamic information contained in sampled data signals. An optimization problem over a set of finite impulse response filters is posed which, while not convex, can provide good quality attributes for classification of signal sources. Characteristics of the optimization problem, possible methods of its solution, and results using nonlinear programming are discussed.<<ETX>>
[1] J.R.B. Cockett,et al. Inductive inference of model structure using hypothesis feedback , 1988, 1988., IEEE International Symposium on Circuits and Systems.
[2] Michael A. Saunders,et al. User''s guide for NPSOL (Ver-sion 4.0): A FORTRAN package for nonlinear programming , 1984 .