Learning Earthquake Wave Arrival Time Picking from Labels with Inaccuracies
For seismologists and geoscientists, LaNCoR reduces the impact of label noise in training data, which is a common and costly problem in supervised learning for seismic signal processing.
The paper introduces LaNCoR, a method to handle noisy labels in seismic P-phase arrival-time picking, achieving up to 28.8% improvement in performance metrics without requiring large-scale training datasets.
Inaccurately labeled training data, or "label noise", poses a significant threat to the integrity of supervised machine learning models. This corruption directly degrades performance by teaching the model erroneous mappings between features and labels, which leads to poor generalization and reduced accuracy on properly labeled validation and test data. Current seismological applications mainly rely on large-scale training sets or data augmentation to reduce the label-noise impact, which can be labor-intensive and costly. Here, we introduce a Label Noise-Contrastive Robust Learning (LaNCoR) approach that can effectively handle noisy labels in seismic signal processing tasks, without requiring large-scale training datasets. In this approach, the input waveform feature and label representation distributions are aligned in the feature space to correct mislabeling and reduce its impact on the training process. We present LaNCoR's performance on the task of P-phase arrival-time picking of real microseismic data using two baseline models and training approaches. Our results indicate that LaNCoR can improve performance by up to 28.8% across performance metrics. This approach holds great promise for model training in seismology and geosciences.