Ziyi “Francis” Yin - 尹梓屹
About me
I am a research geophysicist at Occidental Petroleum Corporation’s subsurface innovation lab, mainly working on generative AI solutions for seismic imaging and inversion.
I earned my PhD degree in Computational Science and Engineering at Georgia Institute of Technology in 2024, advised by Prof. Felix J. Herrmann. I earned my BSc degree in Mathematics and Computer Science at Emory University in 2019, advised by Prof. James G. Nagy.
A super high-level summary of my research is high-dimensional Bayesian inference for computationally expensive parameter-to-observable maps. Generally, my research interests include:
- scientific machine learning
- large-scale inverse problems
- high-performance computing
- computational imaging
- numerical modeling and simulation
- uncertainty quantification
- simulation-based inference
- and their application in geophysical inversion
Open Special Issue Call in Geosciences (MDPI Open-Access) on Geophysical Inversion: link
Recent News
Nov, 2024: WISER was accepted for publication in Geophysics!
Oct, 2024: Check out digital shadow for geological carbon storage monitoring, available as a preprint on arXiv.
Sep, 2024: WISE was featured in Geophysics Bright Spot in The Leading Edge.
Aug, 2024: I joined Oxy as a research geophysicist!
Aug, 2024: Time-lapse full-waveform permeability inversion: a feasibility study was published in The Leading Edge!
Jun, 2024: I successfully defended my PhD dissertation! Check out the slides.
Apr, 2024: WISE was published in Geophysics!
Mar, 2024: Check out An Uncertainty-Aware Digital Twin for Geological Carbon Storage, presented at SIAM UQ 2024.
Nov, 2023: I successfully defended my PhD thesis proposal and got promoted to a PhD candidate!
Oct, 2023: Inference of CO2 flow patterns – a feasibility study was accepted in NeurIPS 2023 Workshop - Tackling Climate Change with Machine Learning as a spotlight paper.
Oct, 2023: Solving multiphysics-based inverse problems with learned surrogates and constraints was published in Advanced Modeling and Simulation in Engineering Sciences.
Jun, 2023: Model-Parallel Fourier Neural Operators as Learned Surrogates for Large-Scale Parametric PDEs was accepted in Computers & Geosciences. Learned multiphysics inversion with differentiable programming and machine learning was accepted in The Leading Edge. Coupled physics inversion for geological carbon storage monitoring, 3D seismic survey design by maximizing the spectral gap and Monitoring subsurface CO2 plumes with sequential Bayesian inference were accepted to be presented in 2023 International Meeting for Applied Geoscience and Energy.
May, 2023: I started my internship at Chevron in Houston, TX and will be staying here until early September. Feel free to email me for coffee catch-up!
Apr, 2023: My IMAGE expanded abstract, Learned coupled inversion for carbon sequestration monitoring and forecasting with Fourier neural operators, was selected as 2022 IMAGE’s Student Oral Paper Honorable Mention!
Apr, 2023: Optimized time-lapse acquisition design via spectral gap ratio minimization was accepted for publication at Geophysics.
Mar, 2023: Check our recent progress on CO2 monitoring, Coupled physics inversion for geological carbon storage monitoring.
Mar, 2023: Check our SIAM presentation, The Next Step: Interoperable Domain-Specific Programming.
Jan, 2023: Derisking geological carbon storage from high-resolution time-lapse seismic to explainable leakage detection was published in the January 2023 special section in seismic resolution in The Leading Edge.
Nov, 2022: I gave several presentations at ML4SEISMIC annual meeting.
Aug, 2022: De-risking Carbon Capture and Sequestration with Explainable CO2 Leakage Detection in Time-lapse Seismic Monitoring Images was accepted in AAAI 2022 Fall Symposium: The Role of AI in Responding to Climate Challenges.
Aug, 2022: The time-lapse seismic monitoring of geological carbon storage work in SLIM group was featured in College of Computing at Georgia Tech news.
May, 2022: Several abstracts were accepted and going to be presented at the 2022 International Meeting for Applied Geoscience and Energy Meeting.
Apr, 2022: On behalf of SLIM group, I gave a 2-hour live tutorial on Julia for Geoscience in TRANSFORM 2022 conference. Check the recording on youtube here.
Apr, 2022: Model-Parallel Fourier Neural Operators as Learned Surrogates for Large-Scale Parametric PDEs was mentioned in Nvidia’s technical blog post.
Aug, 2021: I started to work as a TA for EAS 4803/8803 Seismic monitoring CO2 storage taught by Prof. Felix J. Herrmann in Spring 2022 semester.
Oct, 2021: I passed my PhD qualify exam!
Sep, 2021: I presented our work on Compressive time-lapse seismic monitoring of carbon storage and sequestration with the joint recovery model in Oral Session TL 1: “Advances and Case Studies”, International Meeting for Applied Geoscience & Energy.
Sep, 2021: I gave a talk at Georgia Tech Geophysics Seminar on “low-cost and robust seismic monitoring of carbon storage and sequestration with the joint recovery model”.
Aug, 2021: I started to work as the head TA for CSE/ISYE 6740 Computational Data Analysis taught by Prof. B. Aditya Prakash in Fall 2021 semester.
Aug, 2021: I received SEG Technical Program Registration grant to register the IMAGE 2021 conference. Thanks SEG!
Oct, 2020: I was promoted to be the President of Society of Exploration Geophysicists (SEG) Student Chapter at Georgia Tech. Check our facebook page here.
Jun, 2020: I received SEG/Chevron Student Leadership Symposium (SLS) travel grant to attend the 2020 SEG Annual Meeting. Thanks SEG and Chevron!