Open AccessRecurrent Neural Networks for Computing the Moore‐Penrose Inverse with Momentum Learning
Author(s) -
Zhang Naimin,
Zhang Ting
Publication year - 2020
Publication title -
chinese journal of electronics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.267
H-Index - 25
eISSN - 2075-5597
pISSN - 1022-4653
DOI - 10.1049/cje.2020.02.005
Subject(s) - convergence (economics) , acceleration , inverse , momentum (technical analysis) , artificial neural network , moore–penrose pseudoinverse , algorithm , computer science , rate of convergence , mathematics , matrix (chemical analysis) , artificial intelligence , key (lock) , physics , classical mechanics , geometry , computer security , materials science , finance , economics , composite material , economic growth
We are concerned with a kind of iterative method for computing the Moore‐Penrose inverse, which can be considered as a discrete‐time form of recurrent neural networks. We study the momentum learning scheme of the method and discuss its semi‐convergence when computing the Moore‐Penrose inverse of a rankdeficient matrix. We prove the semi‐convergence for our new acceleration algorithm and obtain the optimal momentum factor which makes the fastest semi‐convergence. Numerical tests demonstrate the effectiveness of our new acceleration algorithm.