Abstract:
Inequalities for modulus of rational functions\title
The paper presents results on the solution of the one-dimensional Burgers equation and the two-dimensional Poisson
equation followed by model inference and visualization. The solution has been obtained using performance lightweight
neural networks in the developed agile project. The final architecture modularly combines a feed-forward neural network
(FFNN) as an encoder and a Kolmogorov – Arnold neural network (KAN) as a decoder.
Key words:Physics-Informed Neural Networks, Deep Learning, Partial Differential Equations.
Fulltext:
PDF file (2466 kB)