SeisLM: a Foundation Model for Seismic Waveforms

• URL: http://arxiv.org/abs/2410.15765v1
• Date: Mon, 21 Oct 2024 08:24:44 GMT
• Title: SeisLM: a Foundation Model for Seismic Waveforms
• Authors: Tianlin Liu, Jannes Münchmeyer, Laura Laurenti, Chris Marone, Maarten V. de Hoop, Ivan Dokmanić,
• Abstract summary: We introduce the Seismic Language Model (SeisLM), a model designed to analyze seismic waveforms. SeisLM is pretrained on a large collection of open-source seismic datasets using a self-supervised contrastive loss. When fine-tuned, SeisLM excels in seismological tasks like event detection, phase-picking, onset time regression, and foreshock-aftershock classification.
• Score: 9.064547137948743
• License:
• Abstract: We introduce the Seismic Language Model (SeisLM), a foundational model designed to analyze seismic waveforms -- signals generated by Earth's vibrations such as the ones originating from earthquakes. SeisLM is pretrained on a large collection of open-source seismic datasets using a self-supervised contrastive loss, akin to BERT in language modeling. This approach allows the model to learn general seismic waveform patterns from unlabeled data without being tied to specific downstream tasks. When fine-tuned, SeisLM excels in seismological tasks like event detection, phase-picking, onset time regression, and foreshock-aftershock classification. The code has been made publicly available on https://github.com/liutianlin0121/seisLM.

Related papers

• High Resolution Seismic Waveform Generation using Denoising Diffusion [3.5046784866523932]
  This study introduces a novel, efficient, and scalable generative model for high-frequency seismic waveform generation. A spectrogram representation of seismic waveform data is reduced to a lower-dimensional submanifold via an autoencoder. A state-of-the-art diffusion model is trained to generate this latent representation, conditioned on key input parameters.
  arXiv  Detail & Related papers  (2024-10-25T07:01:48Z)
• Learning with Noisy Foundation Models [95.50968225050012]
  This paper is the first work to comprehensively understand and analyze the nature of noise in pre-training datasets. We propose a tuning method (NMTune) to affine the feature space to mitigate the malignant effect of noise and improve generalization.
  arXiv  Detail & Related papers  (2024-03-11T16:22:41Z)
• Generalized Neural Networks for Real-Time Earthquake Early Warning [22.53592578343506]
  We employ a data recombination method to create earthquakes occurring at any location with arbitrary station distributions for neural network training. The trained models can then be applied to various regions with different monitoring setups for earthquake detection and parameter evaluation. Our models reliably report earthquake locations and magnitudes within 4 seconds after the first triggered station, with mean errors of 2.6-6.3 km and 0.05-0.17, respectively.
  arXiv  Detail & Related papers  (2023-12-23T10:45:21Z)
• MAPS: A Noise-Robust Progressive Learning Approach for Source-Free Domain Adaptive Keypoint Detection [76.97324120775475]
  Cross-domain keypoint detection methods always require accessing the source data during adaptation. This paper considers source-free domain adaptive keypoint detection, where only the well-trained source model is provided to the target domain.
  arXiv  Detail & Related papers  (2023-02-09T12:06:08Z)
• Distilling Model Failures as Directions in Latent Space [87.30726685335098]
  We present a scalable method for automatically distilling a model's failure modes. We harness linear classifiers to identify consistent error patterns, and induce a natural representation of these failure modes as directions within the feature space. We demonstrate that this framework allows us to discover and automatically caption challenging subpopulations within the training dataset, and intervene to improve the model's performance on these subpopulations.
  arXiv  Detail & Related papers  (2022-06-29T16:35:24Z)
• StorSeismic: A new paradigm in deep learning for seismic processing [0.0]
  StorSeismic is a framework for seismic data processing. We pre-train seismic data, along with synthetically generated ones, in the self-supervised step. Then, we use the labeled synthetic data to fine-tune the pre-trained network in a supervised fashion to perform various seismic processing tasks.
  arXiv  Detail & Related papers  (2022-04-30T09:55:00Z)
• Machine learning-based porosity estimation from spectral decomposed seismic data [0.0]
  Estimating porosity models via seismic data is challenging due to the signal noise and insufficient resolution of seismic data. We propose a machine learning-based workflow to convert seismic data to porosity models.
  arXiv  Detail & Related papers  (2021-11-23T00:03:38Z)
• Learning Signal-Agnostic Manifolds of Neural Fields [50.066449953522685]
  We leverage neural fields to capture the underlying structure in image, shape, audio and cross-modal audiovisual domains. We show that by walking across the underlying manifold of GEM, we may generate new samples in our signal domains.
  arXiv  Detail & Related papers  (2021-11-11T18:57:40Z)
• Anomaly Detection of Time Series with Smoothness-Inducing Sequential Variational Auto-Encoder [59.69303945834122]
  We present a Smoothness-Inducing Sequential Variational Auto-Encoder (SISVAE) model for robust estimation and anomaly detection of time series. Our model parameterizes mean and variance for each time-stamp with flexible neural networks. We show the effectiveness of our model on both synthetic datasets and public real-world benchmarks.
  arXiv  Detail & Related papers  (2021-02-02T06:15:15Z)
• Spatiotemporal Modeling of Seismic Images for Acoustic Impedance Estimation [12.653673008542155]
  Machine learning-based inversion usually works in a trace-by-trace fashion on seismic data. We propose a deep learning-based seismic inversion workflow that models each seismic trace not only temporally but also spatially.
  arXiv  Detail & Related papers  (2020-06-28T00:19:58Z)
• Semiparametric Bayesian Forecasting of Spatial Earthquake Occurrences [77.68028443709338]
  We propose a fully Bayesian formulation of the Epidemic Type Aftershock Sequence (ETAS) model. The occurrence of the mainshock earthquakes in a geographical region is assumed to follow an inhomogeneous spatial point process.
  arXiv  Detail & Related papers  (2020-02-05T10:11:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.