Advances in Subsurface Data Analytics

<p>Part 1: Traditional Machine Learning Approaches<br>1. User Vs. Machine Seismic Attribute Selection for Unsupervised Machine Learning Techniques: Does Human Insight Provide Better Results Than Statistically Chosen Attributes?<br>2. Relative Performance of Support Vector Machine, Decision Trees, and Random Forest Classifiers for Predicting Production Success in US unconventional Shale Plays</p> <p>Part 2: Deep Learning Approaches<br>3. Recurrent Neural Network: application in facies classification</p> <p>4. Recurrent Neural Network for Seismic Reservoir Characterization<br>5. Application of Convolutional Neural Networks for the Classification of Siliciclastic Core Photographs<br>6. Convolutional Neural Networks for Fault Interpretation – Case Study Examples around the World</p> <p>Part 3: Physics-based Machine Learning Approaches<br>7. Scientific Machine Learning for Improved Seismic Simulation and Inversion<br>8. Prediction of Acoustic Velocities using Machine Learning<br>9. Regularized Elastic Full Waveform Inversion using Deep Learning<br>10. A Holistic Approach to Computing First-arrival Traveltimes using Neural Networks</p> <p>Part 4: New Directions<br>11. Application of Artificial Intelligence to Computational Fluid Dynamics <br></p>