We apply the neural style transfer technique (Gatys et al., 2015) to enrich a simplified prior content distribution with features extracted from a geological reference. We demonstrate an application of this approach by transferring the layered texture from the Marmousi II benchmark model to a distribution generated by random Gaussian field. This work offers another way for the generation of target-specific pseudo-random subsurface models.
This document discusses using citizen science and artificial intelligence to classify galaxies. It begins by providing historical context on the recognition of spiral galaxies. It then discusses how Galaxy Zoo used citizen science to classify over a million galaxies. Recent work has used deep learning models to automatically classify galaxies, which can reproduce volunteer classifications with high accuracy. Future surveys like Euclid will require classifying many more distant galaxies, presenting new challenges.
This document provides a biography for Francesca Samsel, including her education, professional experience, publications, exhibitions, and other activities. Some key points:
- She has a MFA from University of Washington and is currently a research associate at the University of Texas at Austin studying art, science, and visualization collaborations.
- Her past professional experience includes positions at UT El Paso, Los Alamos National Lab, and Fordham University where she researched the contributions of art to science communication and understanding.
- She has published and lectured widely on topics related to color theory, scientific visualization, and art-science collaborations.
- She has had numerous solo exhibitions of her artwork focusing on scientific themes and
A Learned Representation for Artistic StyleMayank Agarwal
1) The authors introduce conditional instance normalization, which allows a single style transfer network to capture multiple artistic styles. This is done by normalizing activations according to style-dependent scaling and shifting parameters.
2) A key benefit is the network can stylize an image into different styles with a single forward pass, unlike previous methods that required separate networks for each style.
3) The approach significantly reduces parameters compared to training separate networks, growing linearly with the number of feature maps rather than the number of styles. This also makes adding new styles more efficient.
Overview of accelerated materials design efforts in the Hacking Materials res...Anubhav Jain
This document provides an overview of accelerated materials design efforts in the Hacking Materials research group. It discusses using high-throughput computing and simulations like density functional theory to generate large datasets for materials screening. Machine learning techniques like matminer are used to represent materials as feature vectors to enable predictive modeling. Text mining of scientific literature is also discussed as a way to automatically extract knowledge from millions of published articles to inform new materials discoveries. The goal is to develop automated methods that can suggest the next best computational experiments to optimize properties of interest.
This document discusses generative adversarial networks (GANs) and their applications. It first introduces GANs, describing the generator, discriminator, and objective function. It then discusses various improvements to GANs such as Wasserstein GANs and mode regularization. The document also summarizes applications of GANs to image-to-image translation using CycleGAN and StarGAN, as well as applications to video, speech synthesis, and voice conversion.
This document summarizes a thesis by Alice Gillespie that developed an artificial neural network model called NARX to forecast short-term sea surface elevation for applications in wave energy conversion. The NARX model predicts wave-by-wave surface elevation time series based on previous observations alone. Various preprocessing techniques were investigated as inputs to improve forecast accuracy, including filtering the input data in time and frequency domains and using wavelet transformations. The goal was to develop a model that could reliably forecast sea elevation for up to 3 wave periods (approximately 45 seconds) with minimal error accumulation for use in optimizing wave energy converters.
Use of Micro Satellites for Global Connectivity, High Speed Transmission & Da...IJSRED
This document proposes using micro satellites in very low Earth orbit to create a global connectivity network providing high-speed transmission and data analysis capabilities. The network would consist of interconnected cube satellites drifting in a planned orbit. Using high-efficiency thrusters and quantum data transmission, the network could provide low-latency connectivity down to 20-30 milliseconds globally. In addition to connectivity, the constant Earth observation capabilities could allow for weather monitoring, disaster alerts and scientific data analysis. The goals are to provide high-speed internet access globally, including rural areas, as well as accurate weather data and monitoring of natural disasters.
This document discusses using citizen science and artificial intelligence to classify galaxies. It begins by providing historical context on the recognition of spiral galaxies. It then discusses how Galaxy Zoo used citizen science to classify over a million galaxies. Recent work has used deep learning models to automatically classify galaxies, which can reproduce volunteer classifications with high accuracy. Future surveys like Euclid will require classifying many more distant galaxies, presenting new challenges.
This document provides a biography for Francesca Samsel, including her education, professional experience, publications, exhibitions, and other activities. Some key points:
- She has a MFA from University of Washington and is currently a research associate at the University of Texas at Austin studying art, science, and visualization collaborations.
- Her past professional experience includes positions at UT El Paso, Los Alamos National Lab, and Fordham University where she researched the contributions of art to science communication and understanding.
- She has published and lectured widely on topics related to color theory, scientific visualization, and art-science collaborations.
- She has had numerous solo exhibitions of her artwork focusing on scientific themes and
A Learned Representation for Artistic StyleMayank Agarwal
1) The authors introduce conditional instance normalization, which allows a single style transfer network to capture multiple artistic styles. This is done by normalizing activations according to style-dependent scaling and shifting parameters.
2) A key benefit is the network can stylize an image into different styles with a single forward pass, unlike previous methods that required separate networks for each style.
3) The approach significantly reduces parameters compared to training separate networks, growing linearly with the number of feature maps rather than the number of styles. This also makes adding new styles more efficient.
Overview of accelerated materials design efforts in the Hacking Materials res...Anubhav Jain
This document provides an overview of accelerated materials design efforts in the Hacking Materials research group. It discusses using high-throughput computing and simulations like density functional theory to generate large datasets for materials screening. Machine learning techniques like matminer are used to represent materials as feature vectors to enable predictive modeling. Text mining of scientific literature is also discussed as a way to automatically extract knowledge from millions of published articles to inform new materials discoveries. The goal is to develop automated methods that can suggest the next best computational experiments to optimize properties of interest.
This document discusses generative adversarial networks (GANs) and their applications. It first introduces GANs, describing the generator, discriminator, and objective function. It then discusses various improvements to GANs such as Wasserstein GANs and mode regularization. The document also summarizes applications of GANs to image-to-image translation using CycleGAN and StarGAN, as well as applications to video, speech synthesis, and voice conversion.
This document summarizes a thesis by Alice Gillespie that developed an artificial neural network model called NARX to forecast short-term sea surface elevation for applications in wave energy conversion. The NARX model predicts wave-by-wave surface elevation time series based on previous observations alone. Various preprocessing techniques were investigated as inputs to improve forecast accuracy, including filtering the input data in time and frequency domains and using wavelet transformations. The goal was to develop a model that could reliably forecast sea elevation for up to 3 wave periods (approximately 45 seconds) with minimal error accumulation for use in optimizing wave energy converters.
Use of Micro Satellites for Global Connectivity, High Speed Transmission & Da...IJSRED
This document proposes using micro satellites in very low Earth orbit to create a global connectivity network providing high-speed transmission and data analysis capabilities. The network would consist of interconnected cube satellites drifting in a planned orbit. Using high-efficiency thrusters and quantum data transmission, the network could provide low-latency connectivity down to 20-30 milliseconds globally. In addition to connectivity, the constant Earth observation capabilities could allow for weather monitoring, disaster alerts and scientific data analysis. The goals are to provide high-speed internet access globally, including rural areas, as well as accurate weather data and monitoring of natural disasters.
Surface-related multiple elimination through orthogonal encoding in the laten...Oleg Ovcharenko
We explore the feasibility of surface-related multiple elimination by two-step separation where primaries and multiples are separated in the latent space of a convolutional autoencoder. First, we train a convolutional autoencoder to produce orthogonal embeddings of primaries and multiples. Second, we train another network to classify the latent space embedding of target data into respective wave types and decode predictions back to the data domain. Moreover, we propose an end-to-end workflow for the generation of realistic synthetic seismic data sufficient for knowledge transfer from training on synthetic to inference on field data. We evaluate the two-step separation approach in synthetic setup and highlight the strengths and weaknesses of using masks in encoder latent space for surface-related multiple elimination.
Dual-band generative learning for low-frequency extrapolation in seismic land...Oleg Ovcharenko
The presence of low-frequency energy in seismic data can help mitigate cycle-skipping problems in full-waveform inversion. Unfortunately, the generation and recording of low-frequency signals in seismic exploration remains a non-trivial task. Extrapolation of missing low-frequency content in field data might be addressed in a data-driven framework. In particular, deep learning models trained on synthetic data could be used for inference on the field data. Such an implementation of switching application domains remains challenging. We, therefore, propose the concept of generative dual-band learning to facilitate the knowledge transfer between synthetic and field seismic data applications of low-frequency data extrapolation. We first explain the two-step procedure for training a generative adversarial network (GAN) that extrapolates low frequencies. Then, we describe the workflow for synthetic dataset generation. Finally, we explore the feasibility of the dual-band learning concept on real near-surface land data acquired in Saudi Arabia.
Data-driven methods for the initialization of full-waveform inversionOleg Ovcharenko
Here are the slides from the Ph.D. thesis defense by Oleg Ovcharenko that took place at KAUST in Thuwal, Saudi Arabia. The research is related to deep learning applications for initial model building for FWI and low-frequency extrapolation in seismic data.
Neural network-based low-frequency data extrapolationOleg Ovcharenko
Slides for my talk at SEG Workshop in Manama, Bahrain, December 2017. We introduce an approach to extrapolate for missing low-frequency data from frequency representation of multi-offset seismic data. Meaning that data on multiple high-frequencies is used to infer a single low-frequency for each receiver. In the end, we demonstrate a preliminary example of building an initial model for FWI from the extrapolated data.
Feasibility of moment tensor inversion for a single-well microseismic data us...Oleg Ovcharenko
Slides for my talk at GEO 2018 in Manama, Bahrain. We approach the problem of full moment tensor reconstruction when given data from a single well. Only 5 of 6 moment tensor components are resolved in isotropic medium (in anisotropic all 6 might be resolved in theory), whereas the 6th one is only approximated. We propose a data-driven approach to build a representation of all 6 components from amplitudes of first arrivals at 3-component geophone using a vanilla feed-forward multilayer perceptron.
Slides for my talk at EAGE 2017 in Paris, France. We introduce a technique which identifies corrupted areas in the updates of full-waveform inversions. These areas are then corrected to mitigate cycle-skipping artifacts and thus inversion converges to a more realistic model of the subsurface.
Transfer learning for low frequency extrapolation from shot gathers for FWI a...Oleg Ovcharenko
Slides for my talk at EAGE 2019 in London this June. We attempt to extrapolate for missing low-frequency content in seismic data using a deep learning (DL) approach. We generate a set of random subsurface models and use those to produce a synthetic training dataset. We train a supervised DL model to infer a mono-frequency representation of a common shot gather, given respective data on multiple high frequencies. In the end, we show an example of FWI on extrapolated synthetic data and an example of bandwidth extrapolation on a single shot from field data.
The use of Nauplii and metanauplii artemia in aquaculture (brine shrimp).pptxMAGOTI ERNEST
Although Artemia has been known to man for centuries, its use as a food for the culture of larval organisms apparently began only in the 1930s, when several investigators found that it made an excellent food for newly hatched fish larvae (Litvinenko et al., 2023). As aquaculture developed in the 1960s and ‘70s, the use of Artemia also became more widespread, due both to its convenience and to its nutritional value for larval organisms (Arenas-Pardo et al., 2024). The fact that Artemia dormant cysts can be stored for long periods in cans, and then used as an off-the-shelf food requiring only 24 h of incubation makes them the most convenient, least labor-intensive, live food available for aquaculture (Sorgeloos & Roubach, 2021). The nutritional value of Artemia, especially for marine organisms, is not constant, but varies both geographically and temporally. During the last decade, however, both the causes of Artemia nutritional variability and methods to improve poorquality Artemia have been identified (Loufi et al., 2024).
Brine shrimp (Artemia spp.) are used in marine aquaculture worldwide. Annually, more than 2,000 metric tons of dry cysts are used for cultivation of fish, crustacean, and shellfish larva. Brine shrimp are important to aquaculture because newly hatched brine shrimp nauplii (larvae) provide a food source for many fish fry (Mozanzadeh et al., 2021). Culture and harvesting of brine shrimp eggs represents another aspect of the aquaculture industry. Nauplii and metanauplii of Artemia, commonly known as brine shrimp, play a crucial role in aquaculture due to their nutritional value and suitability as live feed for many aquatic species, particularly in larval stages (Sorgeloos & Roubach, 2021).
Travis Hills' Endeavors in Minnesota: Fostering Environmental and Economic Pr...Travis Hills MN
Travis Hills of Minnesota developed a method to convert waste into high-value dry fertilizer, significantly enriching soil quality. By providing farmers with a valuable resource derived from waste, Travis Hills helps enhance farm profitability while promoting environmental stewardship. Travis Hills' sustainable practices lead to cost savings and increased revenue for farmers by improving resource efficiency and reducing waste.
Comparing Evolved Extractive Text Summary Scores of Bidirectional Encoder Rep...University of Maribor
Slides from:
11th International Conference on Electrical, Electronics and Computer Engineering (IcETRAN), Niš, 3-6 June 2024
Track: Artificial Intelligence
https://www.etran.rs/2024/en/home-english/
ESR spectroscopy in liquid food and beverages.pptxPRIYANKA PATEL
With increasing population, people need to rely on packaged food stuffs. Packaging of food materials requires the preservation of food. There are various methods for the treatment of food to preserve them and irradiation treatment of food is one of them. It is the most common and the most harmless method for the food preservation as it does not alter the necessary micronutrients of food materials. Although irradiated food doesn’t cause any harm to the human health but still the quality assessment of food is required to provide consumers with necessary information about the food. ESR spectroscopy is the most sophisticated way to investigate the quality of the food and the free radicals induced during the processing of the food. ESR spin trapping technique is useful for the detection of highly unstable radicals in the food. The antioxidant capability of liquid food and beverages in mainly performed by spin trapping technique.
Or: Beyond linear.
Abstract: Equivariant neural networks are neural networks that incorporate symmetries. The nonlinear activation functions in these networks result in interesting nonlinear equivariant maps between simple representations, and motivate the key player of this talk: piecewise linear representation theory.
Disclaimer: No one is perfect, so please mind that there might be mistakes and typos.
dtubbenhauer@gmail.com
Corrected slides: dtubbenhauer.com/talks.html
When I was asked to give a companion lecture in support of ‘The Philosophy of Science’ (https://shorturl.at/4pUXz) I decided not to walk through the detail of the many methodologies in order of use. Instead, I chose to employ a long standing, and ongoing, scientific development as an exemplar. And so, I chose the ever evolving story of Thermodynamics as a scientific investigation at its best.
Conducted over a period of >200 years, Thermodynamics R&D, and application, benefitted from the highest levels of professionalism, collaboration, and technical thoroughness. New layers of application, methodology, and practice were made possible by the progressive advance of technology. In turn, this has seen measurement and modelling accuracy continually improved at a micro and macro level.
Perhaps most importantly, Thermodynamics rapidly became a primary tool in the advance of applied science/engineering/technology, spanning micro-tech, to aerospace and cosmology. I can think of no better a story to illustrate the breadth of scientific methodologies and applications at their best.
ANAMOLOUS SECONDARY GROWTH IN DICOT ROOTS.pptxRASHMI M G
Abnormal or anomalous secondary growth in plants. It defines secondary growth as an increase in plant girth due to vascular cambium or cork cambium. Anomalous secondary growth does not follow the normal pattern of a single vascular cambium producing xylem internally and phloem externally.
Surface-related multiple elimination through orthogonal encoding in the laten...Oleg Ovcharenko
We explore the feasibility of surface-related multiple elimination by two-step separation where primaries and multiples are separated in the latent space of a convolutional autoencoder. First, we train a convolutional autoencoder to produce orthogonal embeddings of primaries and multiples. Second, we train another network to classify the latent space embedding of target data into respective wave types and decode predictions back to the data domain. Moreover, we propose an end-to-end workflow for the generation of realistic synthetic seismic data sufficient for knowledge transfer from training on synthetic to inference on field data. We evaluate the two-step separation approach in synthetic setup and highlight the strengths and weaknesses of using masks in encoder latent space for surface-related multiple elimination.
Dual-band generative learning for low-frequency extrapolation in seismic land...Oleg Ovcharenko
The presence of low-frequency energy in seismic data can help mitigate cycle-skipping problems in full-waveform inversion. Unfortunately, the generation and recording of low-frequency signals in seismic exploration remains a non-trivial task. Extrapolation of missing low-frequency content in field data might be addressed in a data-driven framework. In particular, deep learning models trained on synthetic data could be used for inference on the field data. Such an implementation of switching application domains remains challenging. We, therefore, propose the concept of generative dual-band learning to facilitate the knowledge transfer between synthetic and field seismic data applications of low-frequency data extrapolation. We first explain the two-step procedure for training a generative adversarial network (GAN) that extrapolates low frequencies. Then, we describe the workflow for synthetic dataset generation. Finally, we explore the feasibility of the dual-band learning concept on real near-surface land data acquired in Saudi Arabia.
Data-driven methods for the initialization of full-waveform inversionOleg Ovcharenko
Here are the slides from the Ph.D. thesis defense by Oleg Ovcharenko that took place at KAUST in Thuwal, Saudi Arabia. The research is related to deep learning applications for initial model building for FWI and low-frequency extrapolation in seismic data.
Neural network-based low-frequency data extrapolationOleg Ovcharenko
Slides for my talk at SEG Workshop in Manama, Bahrain, December 2017. We introduce an approach to extrapolate for missing low-frequency data from frequency representation of multi-offset seismic data. Meaning that data on multiple high-frequencies is used to infer a single low-frequency for each receiver. In the end, we demonstrate a preliminary example of building an initial model for FWI from the extrapolated data.
Feasibility of moment tensor inversion for a single-well microseismic data us...Oleg Ovcharenko
Slides for my talk at GEO 2018 in Manama, Bahrain. We approach the problem of full moment tensor reconstruction when given data from a single well. Only 5 of 6 moment tensor components are resolved in isotropic medium (in anisotropic all 6 might be resolved in theory), whereas the 6th one is only approximated. We propose a data-driven approach to build a representation of all 6 components from amplitudes of first arrivals at 3-component geophone using a vanilla feed-forward multilayer perceptron.
Slides for my talk at EAGE 2017 in Paris, France. We introduce a technique which identifies corrupted areas in the updates of full-waveform inversions. These areas are then corrected to mitigate cycle-skipping artifacts and thus inversion converges to a more realistic model of the subsurface.
Transfer learning for low frequency extrapolation from shot gathers for FWI a...Oleg Ovcharenko
Slides for my talk at EAGE 2019 in London this June. We attempt to extrapolate for missing low-frequency content in seismic data using a deep learning (DL) approach. We generate a set of random subsurface models and use those to produce a synthetic training dataset. We train a supervised DL model to infer a mono-frequency representation of a common shot gather, given respective data on multiple high frequencies. In the end, we show an example of FWI on extrapolated synthetic data and an example of bandwidth extrapolation on a single shot from field data.
The use of Nauplii and metanauplii artemia in aquaculture (brine shrimp).pptxMAGOTI ERNEST
Although Artemia has been known to man for centuries, its use as a food for the culture of larval organisms apparently began only in the 1930s, when several investigators found that it made an excellent food for newly hatched fish larvae (Litvinenko et al., 2023). As aquaculture developed in the 1960s and ‘70s, the use of Artemia also became more widespread, due both to its convenience and to its nutritional value for larval organisms (Arenas-Pardo et al., 2024). The fact that Artemia dormant cysts can be stored for long periods in cans, and then used as an off-the-shelf food requiring only 24 h of incubation makes them the most convenient, least labor-intensive, live food available for aquaculture (Sorgeloos & Roubach, 2021). The nutritional value of Artemia, especially for marine organisms, is not constant, but varies both geographically and temporally. During the last decade, however, both the causes of Artemia nutritional variability and methods to improve poorquality Artemia have been identified (Loufi et al., 2024).
Brine shrimp (Artemia spp.) are used in marine aquaculture worldwide. Annually, more than 2,000 metric tons of dry cysts are used for cultivation of fish, crustacean, and shellfish larva. Brine shrimp are important to aquaculture because newly hatched brine shrimp nauplii (larvae) provide a food source for many fish fry (Mozanzadeh et al., 2021). Culture and harvesting of brine shrimp eggs represents another aspect of the aquaculture industry. Nauplii and metanauplii of Artemia, commonly known as brine shrimp, play a crucial role in aquaculture due to their nutritional value and suitability as live feed for many aquatic species, particularly in larval stages (Sorgeloos & Roubach, 2021).
Travis Hills' Endeavors in Minnesota: Fostering Environmental and Economic Pr...Travis Hills MN
Travis Hills of Minnesota developed a method to convert waste into high-value dry fertilizer, significantly enriching soil quality. By providing farmers with a valuable resource derived from waste, Travis Hills helps enhance farm profitability while promoting environmental stewardship. Travis Hills' sustainable practices lead to cost savings and increased revenue for farmers by improving resource efficiency and reducing waste.
Comparing Evolved Extractive Text Summary Scores of Bidirectional Encoder Rep...University of Maribor
Slides from:
11th International Conference on Electrical, Electronics and Computer Engineering (IcETRAN), Niš, 3-6 June 2024
Track: Artificial Intelligence
https://www.etran.rs/2024/en/home-english/
ESR spectroscopy in liquid food and beverages.pptxPRIYANKA PATEL
With increasing population, people need to rely on packaged food stuffs. Packaging of food materials requires the preservation of food. There are various methods for the treatment of food to preserve them and irradiation treatment of food is one of them. It is the most common and the most harmless method for the food preservation as it does not alter the necessary micronutrients of food materials. Although irradiated food doesn’t cause any harm to the human health but still the quality assessment of food is required to provide consumers with necessary information about the food. ESR spectroscopy is the most sophisticated way to investigate the quality of the food and the free radicals induced during the processing of the food. ESR spin trapping technique is useful for the detection of highly unstable radicals in the food. The antioxidant capability of liquid food and beverages in mainly performed by spin trapping technique.
Or: Beyond linear.
Abstract: Equivariant neural networks are neural networks that incorporate symmetries. The nonlinear activation functions in these networks result in interesting nonlinear equivariant maps between simple representations, and motivate the key player of this talk: piecewise linear representation theory.
Disclaimer: No one is perfect, so please mind that there might be mistakes and typos.
dtubbenhauer@gmail.com
Corrected slides: dtubbenhauer.com/talks.html
When I was asked to give a companion lecture in support of ‘The Philosophy of Science’ (https://shorturl.at/4pUXz) I decided not to walk through the detail of the many methodologies in order of use. Instead, I chose to employ a long standing, and ongoing, scientific development as an exemplar. And so, I chose the ever evolving story of Thermodynamics as a scientific investigation at its best.
Conducted over a period of >200 years, Thermodynamics R&D, and application, benefitted from the highest levels of professionalism, collaboration, and technical thoroughness. New layers of application, methodology, and practice were made possible by the progressive advance of technology. In turn, this has seen measurement and modelling accuracy continually improved at a micro and macro level.
Perhaps most importantly, Thermodynamics rapidly became a primary tool in the advance of applied science/engineering/technology, spanning micro-tech, to aerospace and cosmology. I can think of no better a story to illustrate the breadth of scientific methodologies and applications at their best.
ANAMOLOUS SECONDARY GROWTH IN DICOT ROOTS.pptxRASHMI M G
Abnormal or anomalous secondary growth in plants. It defines secondary growth as an increase in plant girth due to vascular cambium or cork cambium. Anomalous secondary growth does not follow the normal pattern of a single vascular cambium producing xylem internally and phloem externally.
This presentation explores a brief idea about the structural and functional attributes of nucleotides, the structure and function of genetic materials along with the impact of UV rays and pH upon them.
2. {oleg.ovcharenko, vladimir.kazei}@kaust.edu.saStyle transfer for realistic subsurface models
Team 2
Vladimir Kazei,
Post-doctoral Fellow
Oleg Ovcharenko,
PhD student
Tariq Alkhalifah,
Professor
Daniel Peter,
Assistant Professor
KAUST
Saudi Arabia
4. {oleg.ovcharenko, vladimir.kazei}@kaust.edu.saStyle transfer for realistic subsurface models
Community random model generators 4
Deep-learning tomography
Araya-Polo et al., 2017
Deep-learning inversion: a next generation
seismic velocity-model building method
Yang and Ma, 2019
Deep learning Inversion of Seismic Data
Li et al, 2019
Generative Adversarial Networks for Model
Order Reduction in
Seismic Full-Waveform Inversion,
Richardson, 2018
Velocity model building from raw shot
gathers using machine learning
Øye and Dahl, 2019
Stochastic Seismic Waveform Inversion
using Generative Adversarial Networks as
a Geological Prior
Mosser et al, 2018
5. {oleg.ovcharenko, vladimir.kazei}@kaust.edu.saStyle transfer for realistic subsurface models
Backstory — Bandwidth extrapolation 5
“Deep learning for low-frequency extrapolation from multi-offset seismic data”,
Ovcharenko et al., 2019.
GEOPHYSICS
7. {oleg.ovcharenko, vladimir.kazei}@kaust.edu.saStyle transfer for realistic subsurface models
Backstory — Velocity model building 7
“Mapping seismic data cubes to vertical velocity profiles by deep learning:
New full-waveform inversion paradigm?”, Kazei et al., 2019
submitted to GEOPHYSICS https://github.com/vkazei/deeplogs
8. {oleg.ovcharenko, vladimir.kazei}@kaust.edu.saStyle transfer for realistic subsurface models
Backstory — Elastic transform 8
“Mapping seismic data cubes to vertical velocity profiles by deep learning:
New full-waveform inversion paradigm?”, Kazei et al., 2019
submitted to GEOPHYSICS https://github.com/vkazei/deeplogs
10. {oleg.ovcharenko, vladimir.kazei}@kaust.edu.saStyle transfer for realistic subsurface models
Neural style-transfer 10
Gatys, L.A., Ecker, A.S. and Bethge,
M., 2015. A neural algorithm of artistic
style. arXiv preprint arXiv:1508.06576.
Neckarfront in Tubingen, Germany
Der Schrei by Edvard MunchThe Starry Night by Vincent van Gogh
The Shipwreck of the Minotaur
by J.M.W. Turner
22. {oleg.ovcharenko, vladimir.kazei}@kaust.edu.saStyle transfer for realistic subsurface models
22
+
Marmousi II“Salt”
TensorFlow 1.12.0Python 3.6 Keras 2.2.4 Titan V
Computational aspects
Size:
100 x 300 x 3
L-BFGS:
100 iterations
~ 1 sec / iter
Future: fast style transfer by GAN following (Johnson et al., 2016; Ulyanov et al., 2016)
Demo notebook available at https://github.com/ovcharenkoo/geo-style-keras
45. {oleg.ovcharenko, vladimir.kazei}@kaust.edu.saStyle transfer for realistic subsurface models
References 45
Adler, A., Araya-Polo, M. and Poggio, T., 2019, June. Deep Recurrent Architectures for Seismic Tomography. In 81st EAGE Conference and
Exhibition 2019.
Araya-Polo, M., Jennings, J., Adler, A. and Dahlke, T., 2018. Deep-learning tomography. The Leading Edge, 37(1), pp.58-66.
Gatys, L.A., Ecker, A.S. and Bethge, M., 2015. A neural algorithm of artistic style. arXiv preprint arXiv:1508.06576.
Johnson, J., Alahi, A. and Fei-Fei, L., 2016, October. Perceptual losses for real-time style transfer and super-resolution. In European
conference on computer vision (pp. 694-711). Springer, Cham.
Kazei, V., Ovcharenko, O., Zhang, X., Peter, D. & Alkhalifah, T. "Mapping seismic data cubes to vertical velocity profiles by deep learning:
New full-waveform inversion paradigm?", Geophysics, submitted (2019)
Li, S., Liu, B., Ren, Y., Chen, Y., Yang, S., Wang, Y. and Jiang, P., 2019. Deep learning inversion of seismic data. arXiv preprint arXiv:
1901.07733.
Mosser, L., Dubrule, O. and Blunt, M., 2018, November. Stochastic seismic waveform inversion using generative adversarial networks as a
geological prior. In First EAGE/PESGB Workshop Machine Learning.
Ovcharenko, O., Kazei, V., Kalita, M., Peter, D. and Alkhalifah, T.A., 2019. Deep learning for low-frequency extrapolation from multi-offset
seismic data.
Øye, O.K. and Dahl, E.K., 2019, June. Velocity Model Building from Raw Shot Gathers Using Machine Learning. In 81st EAGE Conference
and Exhibition 2019.
Richardson, A., 2018. Generative adversarial networks for model order reduction in seismic full-waveform inversion. arXiv preprint arXiv:
1806.00828.
Simonyan, K. and Zisserman, A., 2014. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
Ulyanov, D., Lebedev, V., Vedaldi, A. and Lempitsky, V.S., 2016, June. Texture Networks: Feed-forward Synthesis of Textures and Stylized
Images. In ICML (Vol. 1, No. 2, p. 4).
Yang, F. and Ma, J., 2019. Deep-learning inversion: a next generation seismic velocity-model building method. Geophysics, 84(4), pp.
1-133.
46. {oleg.ovcharenko, vladimir.kazei}@kaust.edu.saStyle transfer for realistic subsurface models
Simplified priors + Geological models = target-textured datasets
46
Well-log constraints can be incorporated
Demo notebook available at https://github.com/ovcharenkoo/geo-style-keras
Outlook
Test models in low frequency extrapolation and velocity model building
Fast style transfer and Automated parameter selection
+ =
Conclusions