Null rules for the detection of lower regularity of functions

Calabrò, Francesco; Bravo, Daniel; Carissimo, Chiara; Fazio, Emanuele Di; Pasquale, Antonio Di; Eldray, A.A.M.O.; Fabrizi, Christopher; Gerges, J.G.S.; Palazzo, Simone; Wassef, J.F.F.T.

doi:10.1016/j.cam.2019.05.002

Cited by 4 publications

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed method can be further developed by exploring some alternatives, such as the use of optimal activation functions, as proposed in [14], or the selection of the solution of the under-determined system by using the properties of null rules, as explored in [7]. Finally, the application of the proposed ELM collocation method can be extended in various ways.…”

Section: Discussionmentioning

confidence: 99%

Extreme learning machine collocation for the numerical solution of elliptic PDEs with sharp gradients

Calabrò,

Fabiani,

Siettos

2020

Preprint

View full text Add to dashboard Cite

We introduce a new numerical method based on machine learning to approximate the solution of elliptic partial differential equations with collocation using a set of sigmoidal functions. We show that a feedforward neural network with a single hidden layer with sigmoidal functions and fixed, random, internal weights and biases can be used to compute accurately a collocation solution. The choice to fix internal weights and bias leads to the so-called Extreme Learning Machine network. We discuss how to determine the range for both internal weights and biases in order to obtain a good underlining approximating space, and we explore the required number of collocation points. We demonstrate the efficiency of the proposed method with several one-dimensional diffusion-advection-reaction problems that exhibit steep behaviors, such as boundary layers. The boundary conditions are imposed directly as collocation equations. We point out that there is no need of training the network, as the proposed numerical approach results to a linear problem that can be easily solved using least-squares.Numerical results show that the proposed method achieves a good accuracy.Finally, we compare the proposed method with finite differences and point out the significant improvements in terms of computational cost, thus avoiding the time-consuming training phase.

show abstract

Section: Discussionmentioning

confidence: 99%