A Complex Valued Hebbian Learning Algorithm
نویسنده
چکیده
We present a new training rule for a single-layered linear network with complex valued weights and activation levels. This novel network can be used to extract the principal components of a complex valued data set. We also introduce a new training method that reduces the training time of the complex valued as well as of the real valued network. The use of the new network and training algorithm is illustrated with a problem of compressing images represented in the spectral domain.
منابع مشابه
Fiori : on Blind Separation of Complex Sources by Extended Hebbian Learning
|The aim of this Letter is to present a non-linear extension to Sanger's generalized Hebbian learning algorithm for complex-valued data neural processing, which allows for separating mixed independent circular source signals. The proposed generalization relies on an interesting interpretation of non-classical Hebbian learning proposed by Sudjianto and Has-soun for real-valued neural units.
متن کاملIEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS { PART II 1 Extended Hebbian Learning for Blind Separation
| The aim of this work is to present a non-linear extension to Sanger's generalized Hebbian learning algorithm for complex-valued signal processing by neural networks. A possible choice of the involved non-linearity is discussed by recalling the Sudjianto-Hassoun interpretation of non-linear Hebbian learning. An extension of this interpretation to the complex-valued case leads to a Rayleigh non...
متن کاملComplex independent component analysis by nonlinear generalized Hebbian learning with Rayleigh nonlinearity
The aim of this paper is to present a non-linear extension of the Sanger’s Generalized Hebbian Algorithm to the processing of complex-valued data. A possible choice of the involved non-linearity is discussed recalling the Sudjianto-Hassoun interpretation of the nonlinear Hebbian learning. Extension of this interpretation to the complex case leads to a nonlinearity called Rayleigh function, whic...
متن کاملHebbian and error-correction learning for complex-valued neurons
In this paper, we observe some important aspects of Hebbian and error-correction learning rules for complex-valued neurons. These learning rules, which were previously considered for the multi-valued neuron whose inputs and output are located on the unit circle, are generalized for a complex-valued neuron whose inputs and output are arbitrary complex numbers. The Hebbian learning rule is also c...
متن کاملNonlinear Complex-Valued Extensions of Hebbian Learning: An Essay
The Hebbian paradigm is perhaps the best-known unsupervised learning theory in connectionism. It has inspired wide research activity in the artificial neural network field because it embodies some interesting properties such as locality and the capability of being applicable to the basic weight-and-sum structure of neuron models. The plain Hebbian principle, however, also presents some inherent...
متن کامل