A Complex-Valued RTRL Algorithm for Recurrent Neural Networks

Abstract

A complex-valued real-time recurrent learning (CRTRL) algorithm for the class of nonlinear adaptive filters realized as fully connected recurrent neural networks is introduced. The proposed CRTRL is derived for a general complex activation function of a neuron, which makes it suitable for nonlinear adaptive filtering of complex-valued nonlinear and nonstationary signals and complex signals with strong component correlations. In addition, this algorithm is generic and represents a natural extension of the real-valued RTRL. Simulations on benchmark and real-world complex-valued signals support the approach.

Cite

Text

Goh and Mandic. "A Complex-Valued RTRL Algorithm for Recurrent Neural Networks." Neural Computation, 2004. doi:10.1162/0899766042321779

Markdown

[Goh and Mandic. "A Complex-Valued RTRL Algorithm for Recurrent Neural Networks." Neural Computation, 2004.](https://mlanthology.org/neco/2004/goh2004neco-complexvalued/) doi:10.1162/0899766042321779

BibTeX

@article{goh2004neco-complexvalued,
  title     = {{A Complex-Valued RTRL Algorithm for Recurrent Neural Networks}},
  author    = {Goh, Su Lee and Mandic, Danilo P.},
  journal   = {Neural Computation},
  year      = {2004},
  pages     = {2699-2713},
  doi       = {10.1162/0899766042321779},
  volume    = {16},
  url       = {https://mlanthology.org/neco/2004/goh2004neco-complexvalued/}
}