[S-CG71] Driving Solid Earth Science through Machine Learning

[SCG71-02] Improvement of Generalization in Seismic Ground-Motion Surrogate Models using Gradient-informed Spectral Feature Distillation

*Yuto Kuroki¹, Hirotaka Hachiya^1,2, Asaki Iwaki³, Takahiro Maeda³, Naonori Ueda², Hiroyuki Fujiwara³ (1. Graduate School of Systems Engineering, Wakayama University, 2. Center for AIP, RIKEN, 3. National Research Institute for Earth Science and Disaster Resilience)

Keywords:Seismic ground-motion, Surrogate model, Deep Larning, Knowledge distillation