Machine learning and artificial intelligence in geosciences /

Detalles Bibliográficos
Clasificación:Libro Electrónico
Otros Autores: Moseley, Ben, Krischer, Lion
Formato: Electrónico eBook
Idioma:Inglés
Publicado: Cambridge, MA : Academic Press, 2020.
Colección:Advances in geophysics ; v. 61.
Temas:
Earth sciences > Data processing.
Machine learning.
Apprentissage automatique.
Earth sciences > Data processing
Machine learning
Acceso en línea:Texto completo
Tabla de Contenidos:
  • Intro
  • Machine Learning in Geosciences
  • Copyright
  • Contents
  • Contributors
  • Preface
  • References
  • Chapter One: 70 years of machine learning in geoscience in review
  • 1. Historic machine learning in geoscience
  • 1.1. Expert systems to knowledge-driven AI
  • 1.2. Neural networks
  • 1.3. Kriging and Gaussian processes
  • 2. Contemporary machine learning in geoscience
  • 2.1. Modern machine learning tools
  • 2.2. Support-vector machines
  • 2.3. Random forests
  • 2.4. Modern deep learning
  • 2.5. Neural network architectures
  • 2.6. Convolutional neural network architectures
  • 2.7. Generative adversarial networks
  • 2.8. Recurrent neural network architectures
  • 2.9. The state of ML on geoscience
  • References
  • Chapter Two: Machine learning and fault rupture: A review
  • 1. Introduction
  • 2. Machine learning: A shallow dive
  • 2.1. Learning tasks, performance, and experience
  • 2.2. Learning capacity
  • 2.3. Geophysical data
  • 3. Laboratory studies
  • 3.1. Laboratory geodesy
  • 3.2. Laboratory seismology
  • 4. Field studies
  • 4.1. Techniques
  • 4.1.1. Earthquake catalog building
  • 4.1.1.1. Earthquake detection
  • 4.1.1.2. Phase picking and polarity determination
  • 4.1.1.3. Event association
  • 4.1.1.4. Event location
  • 4.1.2. Seismic waveform denoising and enhancing
  • 4.1.3. Tectonic tremor detection
  • 4.1.4. Fault slip inversion
  • 4.1.5. Automatic detection of geodetic deformation
  • 4.2. Applications
  • 4.2.1. Early warning
  • 4.2.2. Induced seismicity
  • 4.2.3. Earthquake catalog forecasting
  • 5. Conclusion
  • Acknowledgments
  • References
  • Chapter Three: Machine learning techniques for fractured media
  • 1. Introduction
  • 2. Preliminaries
  • 2.1. Governing equations
  • 2.2. DFN to graph mappings
  • 2.2.1. Topology map
  • 2.2.2. Pipe-network map
  • 2.2.3. Flow topology graph (FTG) construction
  • 3. Graph as a DFN reduced-order model
  • 4. Pruned DFN as a reduced-order model
  • 4.1. Existence of backbones
  • 5. Machine learning methods for backbone identification
  • 5.1. Fracture-classification: Labeling by FTG membership
  • 5.2. Fracture-classification: Labeling by mass flux
  • 5.3. Path-classification: Labeling by FTG membership
  • 5.3.1. Logistic regression
  • 5.3.1.1. Random forest
  • 6. Further scope for ML in fractured media
  • References
  • Further reading
  • Chapter Four: Seismic signal augmentation to improve generalization of deep neural networks
  • 1. Introduction
  • 2. Benchmark data and training procedure
  • 3. Augmentations
  • 3.1. Random shift
  • 3.2. Superimposing events
  • 3.3. Superposing noise
  • 3.4. False positive noise
  • 3.5. Channel dropout
  • 3.6. Resampling
  • 3.7. Augmentation for synthetic data generation
  • 4. Discussion
  • 5. Conclusions
  • Acknowledgments
  • References
  • Chapter Five: Deep generator priors for Bayesian seismic inversion
  • 1. Introduction
  • 2. Methodology
  • 2.1. Bayesian inference

Ejemplares similares