Poster #034, Seismology
EikoNet: Solving the Eikonal equation with Deep Neural Networks, with applications to earthquake location
Poster Image:

he method exploits the differentiability of neural networks to calculate the spatial gradients analytically, meaning the network can be trained on its own without ever needing solutions from a finite difference algorithm. Training and inference are highly parallelized, making the approach well-suited for GPUs. EikoNet has low memory overhead, and further avoids the need for travel-time lookup tables
EikoNet is rigorously tested on several toy velocity models, real-world velocity models and sampling methods to demonstrate robustness and versatility. We outlinethe application to earthquake hypocenter inversion, ray multi-pathing, and tomographic modelling. Demonstrating the advantages of this method over conventional finite-difference approaches.
SHOW MORE
EikoNet is rigorously tested on several toy velocity models, real-world velocity models and sampling methods to demonstrate robustness and versatility. We outlinethe application to earthquake hypocenter inversion, ray multi-pathing, and tomographic modelling. Demonstrating the advantages of this method over conventional finite-difference approaches.
SHOW MORE