Intelligent Computational Algorithms Based on Neural Networks: A Survey

ABSTRACT Neural network (NN)‐based intelligent computational algorithms have emerged as a powerful paradigm for addressing complex time‐varying problems in modern autonomous and engineering systems. Unlike conventional learning models that heavily rely on large datasets and offline training, NN‐based algorithms offer a mathematically tractable, real‐time adaptive, and structurally interpretable alternative for intelligent computation. Building upon the background and classification of neural networks, this article systematically examines representative algorithmic models and highlights their practical relevance through analysis of implementation strategies and deployment in intelligent systems. Finally, by presenting illustrative applications in robotics, optimization, multi‐agent coordination, and time‐varying problem solving, this article offers theoretical insights and methodological guidance for the development of next‐generation intelligent computational frameworks.