Continuous Proteretic Hopfield Neural Network in Walsh-based Distributed Storage

A neural network is a series of algorithms that endeavour to recognize underlying relationships in a set of data through a process that mimics the operations of a human brain. It may have uses in many different real-life applications, such as: speech and voice recognition, eCommerce, cybersecurity, and others. The network convergence time is the one of the most important parts of neural networks, which affects the performance of neural network applications. Convergence of the neural network contributes to the process of determining the optimal number of training iterations required to produce the fewest number of errors. In this paper, a specific method based on the mathematical property found in physical systems called "proteretic" is presented. Three learning methods (standard Hopfield, hysteretic Hopfield, and modified proteretic Hopfield) are applied to the Walsh-based distributed memory application. It mathematically and practically demonstrated and approved that using the modified proteretic method causes the network to reach convergence faster than other methods. It’s approved that using the proteretic property with the Walsh-based memory enhances the performance of the storage by accelerating the network's convergence relative to other neural network operations.