Read the original document by opening this link in a new tab.
Table of Contents
Abstract —Artificial neural networks (ANNs) based machine learning models and especially deep learning models have been widely applied in computer vision, signal processing, wireless communications, and many other domains... I. I NTRODUCTION... II. B ACKGROUND... A. Historical Notes... B. Why Complex-Valued Neural Networks... III. A CTIVATION FUNCTIONS OF CVNN S... IV. ...
Summary
Artificial neural networks (ANNs) based machine learning models and especially deep learning models have been widely applied in various domains. This paper surveys the recent development of Complex-Valued Neural Networks (CVNNs) and discusses their potential advantages over real-valued counterparts. It reviews various CVNNs in terms of activation functions, learning and optimization methods, input and output representations, and applications in areas such as signal processing and computer vision. The paper also addresses challenges and future research directions in the field. With a focus on the importance of phase and magnitude in data representation, the discussion highlights the advantages of using CVNNs, which process information using complex-valued parameters and variables. The paper provides a systematic review and categorization of state-of-the-art CVNNs and offers insights into their similarities, differences, advantages, and limitations. It also explores various applications of CVNNs and presents potential research directions. Overall, this work aims to provide a comprehensive overview of complex-valued neural networks.