Resources

Week 1: What is Machine Learning?

Initial meeting to consider logistical meeting concerns, gauge interest in which subjects to cover, and generate a list of ptential invited speakers.

https://cloud.withgoogle.com/build/data-analytics/explore-history-machine-learning/

https://manning-content.s3.amazonaws.com/download/4/02b02cf-c1ac-47ae-8ffd-25f85daad877/SampleCh01.pdf (including a summary of upcoming subjects)

http://www.deeplearningbook.org/contents/intro.html

Week 2: What does it mean to learn?

Readings:

Deep Learning, Goodfellow et al., Chapter 5, sections 5.0 through 5.4 (inclusive)
- 5.1 Learning Algorithms
- 5.2 Capacity, Overfitting and Underfitting
- 5.3 Hyperparameters and Validation Sets
- 5.4 Estimators, Bias and Variance

Strongly suggested reading:

Deep Learning, Goodfellow et al., Chapters 2, 3, 4 (for background Mathematics reviews) https://github.com/janishar/mit-deep-learning-book-pdf/tree/master/chapter-wise-pdf
2: Linear Algebra
3: Probability and Information Theory
4: Numerical Computation
If you don’t want to review a lot of math, at least read Chapter 4.

Additional resources:

If you find the Ch 5 reading too dense, here are three other chapters that cover many of the same concepts without much vector math:

Week 3: Neural nets and deep learning

Readings:

Using neural nets to recognize handwritten digits http://neuralnetworksanddeeplearning.com/chap1.html

Additional readings:

Deep Learning, Goodfellow et al., Chapter 6
http://ciml.info/dl/v0_99/ciml-v0_99-ch04.pdf

http://playground.tensorflow.org/

Week 4: Recurrence, Convolution and GAN

Readings:

Recurrence NN: http://colah.github.io/posts/2015-08-Understanding-LSTMs/
Convolutional NN: https://ujjwalkarn.me/2016/08/11/intuitive-explanation-convnets/
original paper proposing GAN from Goodfellow: https://arxiv.org/pdf/1406.2661.pdf

Additional readings:

RNN detailed information (Goodfellow textbook, chap 10) https://github.com/janishar/mit-deep-learning-book-pdf/blob/master/chapter-wise-pdf/%5B15%5Dpart-2-chapter-10.pdf
CNN detailed information (Goodfellow textbook, chap 9): https://github.com/janishar/mit-deep-learning-book-pdf/blob/master/chapter-wise-pdf/%5B14%5Dpart-2-chapter-9.pdf
Cool activity for CNN: http://scs.ryerson.ca/~aharley/vis/conv/flat.html

Week 5: Examples in Geoscience: Atmospheric Science

Readings:

Rasp et al., 2018. Deep Learning to represent subgrid processes in climate models. PNAS. https://doi.org/10.1073/pnas.1810286115
Liu et al., 2016. Application of Deep Convolutional Neural Networks for Detecting Extreme Weather in Climate Datasets. https://arxiv.org/abs/1605.01156
See list of additional optional readings below

Week 6: Examples in Geoscience: Geophysics

Readings:

Ping et al., 2018. Using generative adversarial networks to improve deep-learning fault interpretation networks. The Leading Edge. https://doi.org/10.1190/tle37080578.1.

Week 7: Examples in Geoscience: Surface Processes

Readings:

Isikdogan et al., 2018. Learning a River Network Extractor Using an Adaptive Loss Function. https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8319927
Palafox et al., 2017. Automated detection of geological landforms on Mars using Convolutional Neural Networks. Computers and Geoscience. https://www.sciencedirect.com/science/article/pii/S0098300416305532

Week 8: Examples in Geoscience: Geophysics

Readings:

Lu et al., 2018. Using generative adversarial networks to improve deep-learning fault interpretation networks. The Leading Edge. https://doi.org/10.1190/tle37080578.1.

Week 9: Guest Speaker -- Mark Lambert

Mark is an image data scientist with the Earth Science and Remote Sensing (ESRS) Unit within the Astromaterials Research and Exploration Science (ARES) Division at the Johnson Space Center. He uses deep learning techniques to improve the usefulness of astronaut photography through automated feature recognition and image segmentation.

Week 10: Guest Speaker -- Paige Bailey

Paige from Microsoft Azure (now at Google Tensorflow team) works as a senior software engineer at Google Brain specializing in AI/ML (TensorFlow). Prior to joining Google, she was a senior software engineer at Microsoft, and a data scientist engineer at Chevron. Paige has over a decade of experience conducting data analysis and predictive modeling using Python.

Week 11: Guest Speaker -- Mic Faragalli

Mic is the Space Exploration and Advanced Technologies Manager at Mission Control Space Services (http://missioncontrolspaceservices.com/) , a startup located on the Carleton University campus. He is also an Adjunct Research Professor in the Department of Mechanical & Aerospace Engineering at Carleton University.

Week 12: Group presentations I

Presenters:

Nur Schuba: Estimating upper mantle velocities using neural networks
Andrew Moodie: StratGAN: synthetic realizations of experimental stratigraphy with generative adversarial networks and image quilting

Week 13: Group presentations II

Presenters:

Harsha Vora: Predicting Onset of Shear Fracture Growth using Critical-Point Indicators and Machine Learning
Zhaoying Li: Using Machine Learning to Recognize and Detect River Plumes
Brandee Carlson: Classifying Sediment Storage and Release Regimes on Experimental Fan Deltas using a Convolutional Neural Network

Additional Information / resources:

Ian Goodfellow’s Deep Learning COMPLETE PDF TEXTBOOK:
- https://github.com/janishar/mit-deep-learning-book-pdf/blob/master/complete-book-bookmarked-pdf/deeplearningbook.pdf

Examples in Geoscience: Atmospheric / Climate Science
- Rasp et al., 2018. Deep Learning to represent subgrid processes in climate models. PNAS. https://doi.org/10.1073/pnas.1810286115
- Klein et al., 2015. A Dynamic Convolutional Layer for short range weather prediction. IEEE proceedings. https://ieeexplore.ieee.org/document/7299117/
- Liu et al., 2016. Application of Deep Convolutional Neural Networks for Detecting Extreme Weather in Climate Datasets. https://arxiv.org/abs/1605.01156
- (review paper) Karpatne et al., 2017. Machine Learning for the Geosciences: Challenges and Opportunities. https://arxiv.org/abs/1711.04708
- Abbot and Marohasy 2017. The application of machine learning for evaluating anthropogenic versus natural climate change. GeoResJ. https://doi.org/10.1016/j.grj.2017.08.001
- Deo and Sahin, 2015. Application of the extreme learning machine algorithm for theprediction of monthly Effective Drought Index in eastern Australia. Atmospheric Research. https://doi.org/10.1016/j.atmosres.2014.10.016
- Ritaban et al., 2013. Deep cognitive imaging systems enable estimation of continental-scale fire incidence from climate data. Scientific Reports. https://www.nature.com/articles/srep03188.
- Horton et al., 2015. Contribution of changes in atmospheric circulation patterns to extreme temperature trends. Nature Letters. https://www.nature.com/articles/nature14550
- Singh et al., 2016. Recent amplification of the North American winter temperature dipole. JGR Atmospheres. https://doi.org/10.1002/2016JD025116

Examples in Geoscience: Remote Sensing
- Lary et al., Machine learning in geosciences and remote sensing. Geoscience Frontiers, 7(1), 3–10. http://doi.org/10.1016/j.gsf.2015.07.003
- (review paper) Zhang et al., 2016. Deep Learning for Remote Sensing Data: A Technical Tutorial on the State of the Art. https://ieeexplore.ieee.org/document/7486259/
- Isikdogan et al., 2017. Surface Water Mapping by Deep Learning. https://ieeexplore.ieee.org/document/8013683
- Isikdogan et al., 2018. Learning a River Network Extractor Using an Adaptive Loss Function. https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8319927

Examples in Geoscience: Seismology
- Zhu and Beroza, 2018. PhaseNet: A Deep-Neural-Network-Based Seismic Arrival Time Picking Method. Submitted to Geophys. J. Int. https://arxiv.org/abs/1803.03211
- Rouet-Leduc et al., 2017. Machine Learning Predicts Laboratory Earthquakes. Geophysical Research Letters. https://doi.org/10.1002/2017GL074677
- Rouet-Leduc et al., 2018. Estimating Fault Friction From Seismic Signals in the Laboratory. Geophysical Research Letters. https://doi.org/10.1002/2017GL076708
- Perol et al., 2018. Convolutional neural network for earthquake detection and location. http://advances.sciencemag.org/content/4/2/e1700578

Examples in Geoscience: Geophysics
- Lu et al., 2018. Using generative adversarial networks to improve deep-learning fault interpretation networks. The Leading Edge. https://doi.org/10.1190/tle37080578.1.

Examples in Geoscience: Surface Processes
- Dupont et al., 2018. Generating Realistic Geology Conditioned on Physical Measurements with Generative Adversarial Networks. https://arxiv.org/abs/1802.03065
- Shu et al., 2018. An automatic methodology for analyzing sorting level of rock particles. Computers and Geoscience. https://doi.org/10.1016/j.cageo.2018.08.001

Examples in Geoscience: Mars / Planetary exploration
- Palafox et al., 2017. Automated detection of geological landforms on Mars using Convolutional Neural Networks. Computers and Geoscience. https://www.sciencedirect.com/science/article/pii/S0098300416305532
- https://software.nasa.gov/software/NPO-46876-1
- https://www.jpl.nasa.gov/news/news.php?feature=6575

Examples in Geoscience: Geochemistry

Examples in Geoscience: Geological Mapping / Well logging / Facies interp.
- Cracknell, M. J., & Reading, A. M. (2014). Geological mapping using remote sensing data: A comparison of five machine learning algorithms, their response to variations in the spatial distribution of training data and the use of explicit spatial information. Computers & Geosciences, 63, 22–33. http://doi.org/10.1016/J.CAGEO.2013.10.008
- Dubois et al., 2006. Comparison of four approaches to a rock facies classification problem. Computers and Geoscience. https://www.sciencedirect.com/science/article/pii/S0098300406001956?via%3Dihub
- https://geodeepdive.org/
- https://macrostrat.org/
- https://rockd.org/
- https://www.strabospot.org/

Examples in Geoscience: Miscellaneous
- Mosser et al., 2017. Reconstruction of three-dimensional porous media using generative adversarial neural networks. https://arxiv.org/abs/1704.03225

Google Sites

Report abuse