AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

$\mathbf{\mathbb{E}^{FWI}}$: Multiparameter Benchmark Datasets for Elastic Full Waveform Inversion of Geophysical Properties

Shihang Feng, Hanchen Wang, Chengyuan Deng, Yinan Feng, Yanhua Liu, Min Zhu, Peng Jin, Yinpeng Chen, Youzuo Lin

NeurIPS 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Machine learning for physical simulation · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
predict Vp and Vs independently with two InversionNets instead of jointly (ablation)InversionNet · EFWI E^FVAPr SSIM: 2% less (l1), 6% less (l2)Hurt
predict Vp and Vs independently instead of jointly (ablation)InversionNet · EFWI E^FVAPr MAE and MSE: 60%, 72%Hurt
test Vp-only InversionNet on data with altered Vs structure (ablation)InversionNet · EFWI E^FVAMAE, RMSE, SSIM: 343% MAE, 296% RMSE, 7% SSIM dropHurt
test Vs-only InversionNet on data with altered Vp structure (l1) (ablation)InversionNet · EFWI E^CFBMAE, RMSE, SSIM: 120% MAE, 89% RMSE, 7% SSIM dropHurt
test Vs-only InversionNet on data with altered Vp structure (l2) (ablation)InversionNet · EFWI E^CFBMAE, RMSE, SSIM: 115% MAE, 88% RMSE, 7% SSIM dropHurt
move from version A to harder version B datasetElasticNet · EFWI Vel familyVp MAE (l1): 0.0308 -> 0.0668Hurt
move from version A to harder version B datasetElasticNet · EFWI CurveFault familyVp MAE (l1): 0.0551 -> 0.1535Hurt
use l2 loss instead of l1 loss (ablation)ElasticNet · EFWI E^FVAVp MAE: 0.0308 -> 0.0235Improved
use l2 loss instead of l1 loss (ablation)ElasticNet · EFWI E^FVBVp MAE: 0.0668 -> 0.1016Hurt
use l2 loss instead of l1 loss (ablation)ElasticNet · EFWI E^CVBVp MAE: 0.1722 -> 0.1682No clear effect
use l2 loss instead of l1 loss (ablation)ElasticNet · EFWI E^CFBPr SSIM: 0.3576 -> 0.4490Improved
use ElasticTransformer instead of ElasticNet (training cost)ElasticNet · EFWI E^Vel familytraining time per epoch: 4m15s -> 1m15sImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy