Geodesic
Optimization of the training dataset for NDM-net (Numerical Dispersion Mitigation neural network)
Numerical methods and programming, Tome 25 (2024) no. 2, pp. 155-174.

Voir la notice de l'article provenant de la source Math-Net.Ru

In this paper we present a new approach to building a training dataset for the NDMnet (Numerical Dispersion Mitigation neural network) neural network, which suppresses numerical dispersion in modeling seismic wave fields. The NDM-net is trained to display the solution of the system of equations of the dynamic theory of elasticity, calculated on a coarse grid, into a solution modeled on a fine grid. However, in order to train an NDM-net, it is necessary to pre-calculate seismograms on a fine grid, which is a time-consuming procedure. To reduce the computational costs of the algorithm, an original approach is proposed that allows to reduce the learning time without loss of accuracy. It is proposed to consider a linear combination of three different parameters: the distance between sources, the similarity of seismograms and the similarity of velocity models as an effective metric for generating a training dataset. The weights of the linear combination are determined using a global sensitivity analysis.
Keywords: numerical dispersion, seismic modelling, deep learning. 
@article{VMP_2024_25_2_a2,
     author = {E. A. Gondyul and V. V. Lisitsa and K. G. Gadylshin and D. M. Vishnevskii},
     title = {Optimization of the training dataset for {NDM-net} {(Numerical} {Dispersion} {Mitigation} neural network)},
     journal = {Numerical methods and programming},
     pages = {155--174},
     publisher = {mathdoc},
     volume = {25},
     number = {2},
     year = {2024},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/VMP_2024_25_2_a2/}
}
TY  - JOUR
AU  - E. A. Gondyul
AU  - V. V. Lisitsa
AU  - K. G. Gadylshin
AU  - D. M. Vishnevskii
TI  - Optimization of the training dataset for NDM-net (Numerical Dispersion Mitigation neural network)
JO  - Numerical methods and programming
PY  - 2024
SP  - 155
EP  - 174
VL  - 25
IS  - 2
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/VMP_2024_25_2_a2/
LA  - ru
ID  - VMP_2024_25_2_a2
ER  - 
%0 Journal Article
%A E. A. Gondyul
%A V. V. Lisitsa
%A K. G. Gadylshin
%A D. M. Vishnevskii
%T Optimization of the training dataset for NDM-net (Numerical Dispersion Mitigation neural network)
%J Numerical methods and programming
%D 2024
%P 155-174
%V 25
%N 2
%I mathdoc
%U http://geodesic.mathdoc.fr/item/VMP_2024_25_2_a2/
%G ru
%F VMP_2024_25_2_a2
E. A. Gondyul; V. V. Lisitsa; K. G. Gadylshin; D. M. Vishnevskii. Optimization of the training dataset for NDM-net (Numerical Dispersion Mitigation neural network). Numerical methods and programming, Tome 25 (2024) no. 2, pp. 155-174. http://geodesic.mathdoc.fr/item/VMP_2024_25_2_a2/