New LaNCoR Method Improves AI Seismic Data Accuracy by 28.8%

A new method called Label Noise-Contrastive Robust Learning (LaNCoR) has been developed to address the issue of inaccurate labels in supervised machine learning for seismology. This approach aligns input waveform feature and label representation distributions to correct mislabeling without requiring extensive training datasets. Experiments on P-phase arrival-time picking demonstrated that LaNCoR can improve performance by up to 28.8% across various metrics, offering a promising solution for model training in geosciences. AI

IMPACT This new method could significantly improve the accuracy of AI models used in seismology and geosciences by effectively handling noisy data.

RANK_REASON The cluster contains an academic paper detailing a new machine learning method for a specific scientific domain. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• LaNCoR
• P-phase

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →