Special Issue "Fast Learning of Neural Networks with Application to Big Data Processes"

Description

Fast learning of neural networks, especially applied for big data processes recently has gained wide attention, with successful showcases in different areas such as the classification, prediction, pattern recognition, and identification. There are many different research directions how to realize fast learning for neural networks which have obtained success in various domains such as the incremental learning of neurons and parameters, evolving techniques over time (in single-pass sample-wise update mode), with changing structure allowed, shallow neural networks with one layered structure only (e.g., extreme learning machines), learning elastic memory online, deep learning, unsupervised learning, fast optimization algorithms than pure backpropagation, or boosting of neural networks with small weak learners (small neural networks) combined. This special issue collected seven high quality papers reporting the performance results related to some of the previously mentioned emerging research directions how to realize fast learning for neural networks.

Period	20 Dec 2019