Neural network and numerical methods performance comparison for prey-predator model

Abstract

In several areas, differential equations are used extensively to simulate a wide range of events. The Prey-Predator model, sometimes referred to the Lotka-Volterra equations, was used as an example in this work. On the other hand, occasionally insufficient data is available to build an explicit model for this problem. Therefore, being able to approximate differential equation solutions is important. This paper's primary contribution is the performance comparison between the implicit Euler approach and the neural network method. The outcomes demonstrate that although the neural network approach takes longer to provide an estimate, it consistently produces better estimates than the implicit Euler technique.