Vladimir Kazei
 | |
| PhD | Geophysics |
|---|---|
| PhD university | St. Petersburg State University |
Vladimir Kazei is a Research Scientist at Seismic Wave Analysis Group at KAUST. Vladimir’s research is focused on deep learning applications to seismic inverse problems. He obtained his PhD in Geophysics from St. Petersburg State University in 2016, sponsored by Shell GSI BV and SEG scholarship (2015). Vladimir has more than 30 publications in peer-reviewed international journals and conference proceedings focused on seismic imaging for exploration geophysics.
Vladimir is the winner of NVIDIA-KAUST GPU hackathons 2017 & 2019. He pioneered deep learning-based low-frequency extrapolation for full-waveform inversion in 2017 and released one of the first open-source projects on deep learning-based velocity model building in 2019.
SEG J. Clarence Karcher Award 2023
Vladimir Kazei is a research geophysicist at Aramco Americas where he has received accolades from a wide range of researchers and explorationists for his contributions to distributed acoustic sensing (DAS), machine learning, and full-waveform inversion (FWI). He received his PhD degree from the Saint Petersburg State University and Schmidt Institute of the Physics of the Earth of Russian Academy of Sciences in 2016. He then worked at King Abdullah University of Science and Technology as postdoc and research scientist prior to joining Aramco Americas in 2020.
Kazei has made substantial contributions to seismic inversion using artificial intelligence and machine learning methods. His work in mapping multiple common-midpoint gathers of raw field data into depth-domain vertical velocity profiles was characterized as groundbreaking. He is reported to be the first to successfully use deep learning for low-frequency exploration in geophysics working with Oleg Ovcharenko and others. Kazei’s study of orthorhombic FWI resulted in the first radiation scattering pattern atlas – an ultimate theoretical guide to general anisotropic elastic FWI. Additionally, Kazei’s work in inverting DAS amplitudes using conservation of energy was pinned as a new theoretical contribution paving the way towards instant inversion of density along the boreholes from DAS amplitudes. He has chaired SEG technical sessions, co-organized DAS workshops, coedited a DAS special section in GEOPHYSICS, served as a guest editor of The Leading Edge, is a consistent reviewer for GEOPHYSICS, and has published more than 10 journal articles, receiving more than 700 citations.
Biography Citation for the J. Clarence Karcher Award
by Weichang Li
It is with great pleasure that I am writing this citation for Vladimir Kazei for the J. Clarence Karcher Award. As a young geophysicist, Vladimir has already made outstanding research contributions in several areas including full-waveform inversion (FWI), deep learning seismic applications, and fiber-optic distributed acoustic sensing (DAS) imaging and inversion. This is evidenced by his influential publications, lead roles in steering research in these directions by organizing workshops and editing special sections, and recognitions from the SEG community.
I first met Vladimir in 2018 at the SEG Annual Meeting post-convention workshop on Data Analytics and Machine Learning for Geoscience Applications which I organized and where Vladimir presented his work on FWI with Machine Learning-assisted low frequency extrapolation. We then closely worked together after he joined the Aramco Research Center in Houston in 2020. I think his work in applying deep learning (more specifically, convolutional neural network models) to map multiple common-midpoint gathers in depth-domain vertical velocity profiles was groundbreaking. It was one of the first applications to field data of velocity model building via deep learning that had very promising results. Vladimir and his coauthors also pioneered deep learning for low-frequency extrapolation from multioffset seismic data and developed wavenumber illumination-based theoretical basis. Papers from both works have been well cited (59 and 124) since publication in 2021 and 2019, respectively.
I am very familiar with Vladimir’s work in fiber-optic DAS with applications in vertical seismic profile (VSP) using deep learning techniques. His work has produced a series of papers that were presented at the International Meeting for Applied Geoscience and Energy (IMAGE) and MDPI Sensors that led to both practically relevant and theoretically appealing results for DAS data QC, DAS VSP inversion, and DAS logging. The paper that we coauthored on predicting density and velocity ahead of the bit with zero-offset VSP using deep learning was recognized as being in the top 25 presentations at IMAGE ’21. It is clear from the list of these achievements that Vladimir has already made major contributions to the community of applied geophysics in his young career.
Additionally, Vladimir has been a very active member of the geophysics community. He led and co-organized a DAS postconvention workshop at IMAGE ’21, IMAGE ’22, and IMAGE ’23; chaired and cochaired various DAS and deep learning sessions; and served as guest editor for two special sections in The Leading Edge focused on the digital transformation.
Based on what Vladimir has achieved in his early career and his potential as a future leader in this field, the J. Clarence Karcher Award is a very fitting recognition.
External links
| find literature about Vladimir Kazei |
|
|
|
|
|
|
|