<?xml version="1.0" encoding="UTF-8"?>
<rss  xmlns:atom="http://www.w3.org/2005/Atom" 
      xmlns:media="http://search.yahoo.com/mrss/" 
      xmlns:content="http://purl.org/rss/1.0/modules/content/" 
      xmlns:dc="http://purl.org/dc/elements/1.1/" 
      version="2.0">
<channel>
<title>SEEM</title>
<link>https://seem-kfupm.github.io/blog/</link>
<atom:link href="https://seem-kfupm.github.io/blog/index.xml" rel="self" type="application/rss+xml"/>
<description>Updates from SEEM</description>
<generator>quarto-1.10.18</generator>
<lastBuildDate>Thu, 03 Sep 2026 00:00:00 GMT</lastBuildDate>
<item>
  <title>Ayrat Abdullin defends PhD dissertation on real-time microseismic monitoring with deep learning</title>
  <dc:creator>Ayrat Abdullin</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2026-09-03-ayrat-abdullin-phd-defense.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/blog/2026-09-03-ayrat-abdullin-phd-defense/hero.jpg" class="img-fluid" alt="Ayrat Abdullin presenting his PhD dissertation defense, with a slide summarizing the five papers of his dissertation"></p>
</div>
<p>On September 3, 2026, <strong>Ayrat Abdullin</strong> stood before his committee in the Department of Geosciences, College of Petroleum Engineering &amp; Geosciences, at King Fahd University of Petroleum &amp; Minerals, to defend his PhD dissertation, <em>Towards Real-Time Microseismic Monitoring Using Deep Learning</em>. The dissertation, completed under the supervision of Dr.&nbsp;Umair bin Waheed, closes out several years of work at the intersection of deep learning and subsurface monitoring.</p>
<p>Rather than a single, self-contained thesis, the dissertation is a compilation of five research papers, each addressing a different stage of the same problem: making microseismic monitoring fast, reliable, and usable in real time. Together, they cover the full pipeline — detection, classification, phase picking, source localization, and velocity estimation — each rebuilt around deep learning methods designed to work under real-world constraints on speed and data availability.</p>
<section id="the-five-papers" class="level2">
<h2 class="anchored" data-anchor-id="the-five-papers">The five papers</h2>
<ol type="1">
<li><strong>Explainable detection.</strong> Applies Gradient-weighted Class Activation Mapping (Grad-CAM) and Shapley Additive Explanations (SHAP) to a PhaseNet-style detector, showing that explainability tools can do more than interpret a black-box model — a SHAP-gated version of the network improved robustness to noise, reaching an F1-score of 0.98 on a 9,000-waveform test set versus 0.97 for the baseline.</li>
<li><strong>Lightweight classification.</strong> Introduces a compact Fourier Neural Operator for discriminating genuine microseismic events from noise in continuous data streams, cutting the computational cost that has kept many high-accuracy deep learning detectors out of real-time deployment.</li>
<li><strong>Data-efficient phase picking.</strong> Adapts PhaseNO, a network-wide earthquake phase picker pre-trained on more than 57,000 records, to hydraulic-fracturing microseismic data using parameter-efficient transfer learning — fine-tuning only 3.6% of the model’s parameters on 200 labeled recordings while keeping the accuracy gains of the large pre-trained network.</li>
<li><strong>Physics-informed source localization.</strong> Extends a U-shaped, physics-informed neural operator to distributed acoustic sensing (DAS) data for microseismic source localization, tested on data from the Utah FORGE geothermal field.</li>
<li><strong>Joint velocity and location inversion.</strong> Proposes a conditional Schrödinger bridge formulation to jointly recover subsurface velocity structure and microseismic source locations, connecting the dissertation’s monitoring tools to the broader inversion problem.</li>
</ol>
<p>Taken together, the five papers work toward the same goal: deep learning systems accurate enough for operational monitoring of induced seismicity and reservoir stimulation, without requiring more data or compute than a real monitoring network can provide.</p>
<div class="quarto-figure quarto-figure-center">
<figure class="figure">
<p><img src="https://seem-kfupm.github.io/assets/blog/2026-09-03-ayrat-abdullin-phd-defense/committee.jpg" class="img-fluid figure-img"></p>
<figcaption>Ayrat Abdullin with his PhD dissertation committee after the defense</figcaption>
</figure>
</div>
<p>The dissertation committee included Dr.&nbsp;Umair bin Waheed (Advisor), Prof.&nbsp;Abdullatif Al-Shuhail, Dr.&nbsp;Sherif Mahmoud, Dr.&nbsp;Naveed Iqbal, and Dr.&nbsp;Leo Eisner (External). Congratulations to Ayrat on this achievement.</p>


</section>

 ]]></description>
  <category>Student achievements</category>
  <category>Microseismic monitoring</category>
  <category>Deep learning</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2026-09-03-ayrat-abdullin-phd-defense.html</guid>
  <pubDate>Thu, 03 Sep 2026 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/blog/2026-09-03-ayrat-abdullin-phd-defense/hero.jpg" medium="image" type="image/jpeg"/>
</item>
<item>
  <title>AI-accelerated DAS sensitivity modeling presented in Kuala Lumpur</title>
  <dc:creator>Ayrat Abdullin</dc:creator>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <dc:creator>Dr. Umair bin Waheed</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2026-07-21-das-sensitivity-workshop.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/blog/2026-07-21-das-sensitivity-workshop/hero.jpg" class="img-fluid" alt="Distributed acoustic sensing measurements transformed into a coherent seismic representation"></p>
</div>
<p>SEEM PhD student Ayrat Abdullin presented <em>AI-Accelerated Upscaling of DAS Sensitivity Maps for Faster Fibre-Optic Monitoring-Array Design</em> at the SEG/SPE workshop <em>Fibre for the Future: AI-Driven Intelligence</em> in Kuala Lumpur, Malaysia.</p>
<p>The work, co-authored by Dr.&nbsp;Denis Anikiev and Dr.&nbsp;Umair bin Waheed, evaluates Fourier-neural-operator and residual-convolutional-network workflows for reconstructing high-resolution distributed-acoustic-sensing sensitivity maps from coarse-grid forward simulations.</p>
<p>The learned approaches are designed to accelerate monitoring-array studies while preserving geometry-controlled low-sensitivity regions and localized anomalies more effectively than conventional interpolation. The results support more scalable design and assessment of fibre-optic monitoring configurations.</p>



 ]]></description>
  <category>Student achievements</category>
  <category>Distributed acoustic sensing</category>
  <category>Neural operators</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2026-07-21-das-sensitivity-workshop.html</guid>
  <pubDate>Tue, 21 Jul 2026 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/blog/2026-07-21-das-sensitivity-workshop/hero.jpg" medium="image" type="image/jpeg"/>
</item>
<item>
  <title>GFZ visit advances collaboration on generative subsurface modeling</title>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2026-07-15-gfz-generative-subsurface-seminar.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/illustrations/geological-modeling.svg" class="img-fluid" alt="A geological cross-section produced through an AI-assisted modeling workflow"></p>
</div>
<p>As part of an ongoing research collaboration in Germany, Dr.&nbsp;Denis Anikiev visited the GFZ Helmholtz Centre for Geosciences in Potsdam and delivered the invited seminar <em>Toward Generative Subsurface Models: Diffusion-Based Approaches for Geophysical Inversion and Geological Modeling</em>.</p>
<p>The seminar presented recent work on diffusion-based geophysical inversion, uncertainty-aware subsurface characterization, and integrated geological modeling. Technical discussions during the visit focused on the next generation of GeoSIRR and opportunities to combine generative AI with geological knowledge and reproducible modeling workflows.</p>
<p><a href="../../projects/geosirr.html" class="btn btn-primary">Explore the GeoSIRR project</a></p>



 ]]></description>
  <category>Invited talks</category>
  <category>International collaboration</category>
  <category>Generative AI</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2026-07-15-gfz-generative-subsurface-seminar.html</guid>
  <pubDate>Wed, 15 Jul 2026 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/illustrations/geological-modeling.svg" medium="image" type="image/svg+xml"/>
</item>
<item>
  <title>Explainable PhaseNet paper and reproducible software released</title>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <dc:creator>Dr. Umair bin Waheed</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2026-07-01-explainable-phasenet-release.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/illustrations/explainable-ai.svg" class="img-fluid" alt="Highlighted P- and S-wave regions on a seismic waveform"></p>
</div>
<p>The article <em>Explainable AI for Microseismic Event Detection</em>, led by SEEM PhD student Ayrat Abdullin with Dr.&nbsp;Denis Anikiev and Dr.&nbsp;Umair bin Waheed, is available in <em>Artificial Intelligence in Geosciences</em> <span class="citation" data-cites="abdullin2026xai">(Abdullin et al. 2026)</span>.</p>
<p>The study applies Grad-CAM and SHAP to examine the waveform evidence behind PhaseNet decisions. It also introduces a SHAP-gated inference scheme designed to improve detection reliability and robustness under strong noise.</p>
<p>To support transparent and reproducible research, the authors released the implementation, experiments, and supporting scripts through the SEEM-KFUPM GitHub organization and archived version 1.0 on Zenodo.</p>
<p><a href="../../projects/xai-phasenet.html" class="btn btn-primary">Project story</a> <a href="https://doi.org/10.1016/j.aiig.2026.100246" class="btn btn-outline-primary">Article</a> <a href="https://github.com/SEEM-KFUPM/xAI_PhaseNet" class="btn btn-outline-primary">GitHub repository</a> <a href="https://doi.org/10.5281/zenodo.20999758" class="btn btn-outline-primary">Zenodo archive</a></p>




<div id="quarto-appendix" class="default"><section class="quarto-appendix-contents" id="quarto-bibliography"><h2 class="anchored quarto-appendix-heading">References</h2><div id="refs" class="references csl-bib-body hanging-indent">
<div id="ref-abdullin2026xai" class="csl-entry">
Abdullin, Ayrat, Denis Anikiev, and Umair Bin Waheed. 2026. <span>“Explainable <span>AI</span> for Microseismic Event Detection.”</span> <em>Artificial Intelligence in Geosciences</em> 7 (3): 100246. <a href="https://doi.org/10.1016/j.aiig.2026.100246">https://doi.org/10.1016/j.aiig.2026.100246</a>.
</div>
</div></section></div> ]]></description>
  <category>Research releases</category>
  <category>Explainable AI</category>
  <category>Microseismic monitoring</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2026-07-01-explainable-phasenet-release.html</guid>
  <pubDate>Wed, 01 Jul 2026 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/illustrations/explainable-ai.svg" medium="image" type="image/svg+xml"/>
</item>
<item>
  <title>AI-driven seismic and microseismic workflows presented at Saudi Aramco</title>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2026-05-13-aramco-ai-seismic-seminar.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/illustrations/research-workshop.svg" class="img-fluid" alt="A researcher presenting AI and geoscience concepts to an audience"></p>
</div>
<p>Dr.&nbsp;Denis Anikiev delivered an invited presentation to the Saudi Aramco Geophysical Imaging Department Innovation Team Seminar Series titled <em>From Acquisition to Insight: AI-Driven Seismic Imaging and Microseismic Workflows</em>.</p>
<p>The seminar connected recent advances in seismic imaging and microseismic monitoring with diffusion models, neural operators, deep learning, and physics-informed methods. The examples illustrated how these approaches can support subsurface characterization from data acquisition through interpretation.</p>
<p>The discussion generated interest in potential collaboration between CPG and Saudi Aramco on computational geophysics, seismic imaging, and AI-enabled geophysical workflows.</p>



 ]]></description>
  <category>Invited talks</category>
  <category>Seismic imaging</category>
  <category>Artificial intelligence</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2026-05-13-aramco-ai-seismic-seminar.html</guid>
  <pubDate>Wed, 13 May 2026 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/illustrations/research-workshop.svg" medium="image" type="image/svg+xml"/>
</item>
<item>
  <title>3D density model reveals interaction between the Hainan plume and oceanic lithosphere</title>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2026-05-05-hainan-plume-density-model.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/illustrations/geological-modeling.svg" class="img-fluid" alt="A geological cross-section representing density variations in the lithosphere"></p>
</div>
<p>A new study co-authored by Dr.&nbsp;Denis Anikiev presents a three-dimensional lithospheric density model of the South China Sea and examines how the Hainan mantle plume may have interacted with the oceanic lithosphere.</p>
<p>The model integrates gravity observations with geological and geophysical constraints. The results identify a north–south density and temperature contrast across the extinct spreading axis in the eastern sub-basin, supporting a scenario in which plume material influenced asymmetric seafloor spreading through lateral flow toward the ridge.</p>
<p>Published in <em>Journal of Geophysical Research: Solid Earth</em>, the work is an outcome of collaboration among researchers at Nanjing University, GFZ Helmholtz Centre for Geosciences, KFUPM, and partner institutions.</p>
<p><a href="https://doi.org/10.1029/2025JB031818" class="btn btn-primary">Read the article</a></p>



 ]]></description>
  <category>Publications</category>
  <category>Gravity modeling</category>
  <category>International collaboration</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2026-05-05-hainan-plume-density-model.html</guid>
  <pubDate>Tue, 05 May 2026 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/illustrations/geological-modeling.svg" medium="image" type="image/svg+xml"/>
</item>
<item>
  <title>SEEM runs hands-on workshop on AI coding agents</title>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <dc:creator>Dr. Umair bin Waheed</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2026-04-29-ai-coding-agents-workshop.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/illustrations/research-workshop.svg" class="img-fluid" alt="Presentation, code, and collaborators in a research coding workshop"></p>
</div>
<p>Dr.&nbsp;Denis Anikiev and Dr.&nbsp;Umair bin Waheed, together with students from the SEEM Group, organized a hands-on workshop titled <em>Introduction to AI Coding Agents: From Idea to Reproducible Mini-Projects</em> for students enrolled in a machine-learning course.</p>
<p>Participants used Visual Studio Code and Codex to develop small, structured scientific-code projects. The guided exercises emphasized project organization, debugging, documentation, and reproducibility alongside critical review of AI-generated code and its limitations.</p>
<p>The session was the first step toward an expanded two-day training initiative for PhD students, postdoctoral researchers, and research staff. The public project page brings together the reusable workshop materials and supporting resources.</p>
<p><a href="../../projects/ai-coding-agents-workshop.html" class="btn btn-primary">Project story</a> <a href="https://github.com/SEEM-KFUPM/AI-Coding-Agents-Workshop" class="btn btn-outline-primary">Workshop materials</a></p>



 ]]></description>
  <category>Workshops</category>
  <category>Training</category>
  <category>Reproducibility</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2026-04-29-ai-coding-agents-workshop.html</guid>
  <pubDate>Wed, 29 Apr 2026 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/illustrations/research-workshop.svg" medium="image" type="image/svg+xml"/>
</item>
<item>
  <title>GeoSIRR preprint and open research software released</title>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <dc:creator>Dr. Umair bin Waheed</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2026-03-10-geosirr-preprint-release.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/illustrations/geological-modeling.svg" class="img-fluid" alt="A geological cross-section generated from a conversational prompt"></p>
</div>
<p>The GeoSIRR 1.0 discussion preprint is now available through EGUsphere, accompanied by public source code and an archived software release <span class="citation" data-cites="anikiev2026geosirr">(Anikiev et al. 2026)</span>. The manuscript is under review for <em>Geoscientific Model Development</em>.</p>
<p>GeoSIRR—Geological Section Interpretation, Reconstruction &amp; Refinement—translates geological narratives into a domain-specific language and coordinate-based cross-sections. Automated completeness, geometry, and topology checks support reproducible model generation and conversational refinement.</p>
<p>The release builds on work presented at the 19th ILP Workshop in Riyadh and the 2026 IPTC Summit on AI for the Energy Industry. Together, the preprint, code, and examples make the workflow available for evaluation, education, and further development.</p>
<p><a href="../../projects/geosirr.html" class="btn btn-primary">Project story</a> <a href="https://doi.org/10.5194/egusphere-2025-6545" class="btn btn-outline-primary">Preprint</a> <a href="https://github.com/SEEM-KFUPM/GeoSIRR" class="btn btn-outline-primary">GitHub repository</a> <a href="https://doi.org/10.5281/zenodo.18097054" class="btn btn-outline-primary">Zenodo archive</a></p>




<div id="quarto-appendix" class="default"><section class="quarto-appendix-contents" id="quarto-bibliography"><h2 class="anchored quarto-appendix-heading">References</h2><div id="refs" class="references csl-bib-body hanging-indent">
<div id="ref-anikiev2026geosirr" class="csl-entry">
Anikiev, Denis, Juan Esteban Mosquera, Korhan Ayranci, Judith Bott, and Umair Bin Waheed. 2026. <span>“<span>GeoSIRR</span> 1.0: Conversational Geological Cross-Section Modeling Using Large Language Models.”</span> <em>EGUsphere</em>, ahead of print. <a href="https://doi.org/10.5194/egusphere-2025-6545">https://doi.org/10.5194/egusphere-2025-6545</a>.
</div>
</div></section></div> ]]></description>
  <category>Research releases</category>
  <category>Geological modeling</category>
  <category>Large language models</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2026-03-10-geosirr-preprint-release.html</guid>
  <pubDate>Tue, 10 Mar 2026 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/illustrations/geological-modeling.svg" medium="image" type="image/svg+xml"/>
</item>
<item>
  <title>LLM-based assistants for geoscientific software published in IMAGE proceedings</title>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <dc:creator>Dr. Umair bin Waheed</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2026-02-18-llm-assistants-geoscientific-software.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/illustrations/geological-modeling.svg" class="img-fluid" alt="A conversational prompt producing a structured geological cross-section"></p>
</div>
<p>The expanded abstract <em>Leveraging LLM-based Assistants for Geoscientific Software</em> by Denis Anikiev, Umair bin Waheed, and Korhan Ayranci has been published in the proceedings of the Fifth International Meeting for Applied Geoscience &amp; Energy.</p>
<p>The study evaluates how large language model assistants, supported by retrieval-augmented generation and prompt engineering, can help users work with complex geoscientific software. Using the IGMAS+ 3D modeling platform as a case study, the authors compared assistant responses with expert guidance for both introductory and advanced questions.</p>
<p><a href="https://doi.org/10.1190/image2025-4306506.1" class="btn btn-primary">Read the expanded abstract</a></p>



 ]]></description>
  <category>Publications</category>
  <category>Large language models</category>
  <category>Research software</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2026-02-18-llm-assistants-geoscientific-software.html</guid>
  <pubDate>Wed, 18 Feb 2026 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/illustrations/geological-modeling.svg" medium="image" type="image/svg+xml"/>
</item>
<item>
  <title>Gravity-modeling colloquium connects global expertise with SEEM research</title>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2026-02-11-gravity-modeling-colloquium.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/illustrations/geological-modeling.svg" class="img-fluid" alt="A geological cross-section representing a three-dimensional subsurface model"></p>
</div>
<p>The College of Petroleum Engineering &amp; Geosciences hosted Prof.&nbsp;Hans-Jürgen Götze of Kiel University for a colloquium on advances in gravity modeling and their relevance to subsurface exploration.</p>
<p>The lecture showed how gravity observations can be integrated with seismic and geological information to improve structural interpretation across crustal and reservoir scales. Examples from the development and application of IGMAS+ illustrated the value of constrained 3D modeling for reducing interpretation uncertainty.</p>
<p>During the visit, Prof.&nbsp;Götze also joined SEEM meetings on current gravity-modeling projects and emerging approaches that connect geophysical workflows with generative AI.</p>



 ]]></description>
  <category>Visitors</category>
  <category>Gravity modeling</category>
  <category>International collaboration</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2026-02-11-gravity-modeling-colloquium.html</guid>
  <pubDate>Wed, 11 Feb 2026 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/illustrations/geological-modeling.svg" medium="image" type="image/svg+xml"/>
</item>
<item>
  <title>Generative geophysics seminars at Freie Universität Berlin</title>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2025-12-18-fu-berlin-generative-geophysics-seminars.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/illustrations/research-workshop.svg" class="img-fluid" alt="A researcher presenting AI and coding concepts to an audience"></p>
</div>
<p>During an academic visit hosted by Prof.&nbsp;Dr.&nbsp;Serge Shapiro, Dr.&nbsp;Denis Anikiev delivered two invited seminars at Freie Universität Berlin.</p>
<p>The Geo-Colloquium talk, <em>Emerging Trends in Generative AI for Seismic Imaging</em>, examined applications of generative models in exploration, earthquake-hazard research, and geothermal studies. A second seminar for the Applied and Earthquake Seismology group, <em>Machine Learning and Generative AI for Microseismicity</em>, focused on data-driven methods for passive-seismic analysis.</p>
<p>The visit also provided an opportunity to discuss ongoing cooperation and new directions in AI-driven geophysical imaging.</p>



 ]]></description>
  <category>Invited talks</category>
  <category>International collaboration</category>
  <category>Generative AI</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2025-12-18-fu-berlin-generative-geophysics-seminars.html</guid>
  <pubDate>Thu, 18 Dec 2025 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/illustrations/research-workshop.svg" medium="image" type="image/svg+xml"/>
</item>
<item>
  <title>AI for microseismicity presented to the CSEG Microseismic User Group</title>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2025-12-10-cseg-microseismic-user-group-talk.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/illustrations/seismic-localization.svg" class="img-fluid" alt="Subsurface seismic event localized from a surface monitoring array"></p>
</div>
<p>Dr.&nbsp;Denis Anikiev delivered an invited technical presentation to the Microseismic User Group of the Canadian Society of Exploration Geophysicists titled <em>Machine Learning and Generative AI for Microseismicity: Toward Intelligent, Adaptive, and Interpretable Passive Seismic Workflows</em>.</p>
<p>The presentation reviewed how supervised and unsupervised learning, physics-informed models, neural operators, and generative diffusion methods can support the detection, reconstruction, and localization of weak microseismic signals. It also discussed the role of explainable-AI methods such as Grad-CAM and SHAP in evaluating model behavior.</p>
<p>The talk outlined a path toward passive-seismic workflows that combine data-driven performance with physical constraints and interpretable results.</p>



 ]]></description>
  <category>Invited talks</category>
  <category>Microseismic monitoring</category>
  <category>Artificial intelligence</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2025-12-10-cseg-microseismic-user-group-talk.html</guid>
  <pubDate>Wed, 10 Dec 2025 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/illustrations/seismic-localization.svg" medium="image" type="image/svg+xml"/>
</item>
<item>
  <title>SEEM hosts seminar on generative microseismic localization</title>
  <dc:creator>Dr. Denis Anikiev</dc:creator>
  <dc:creator>Dr. Umair bin Waheed</dc:creator>
  <link>https://seem-kfupm.github.io/blog/posts/2025-11-19-kaust-generative-localization-seminar.html</link>
  <description><![CDATA[ 





<div class="blog-post-hero">
<p><img src="https://seem-kfupm.github.io/assets/illustrations/seismic-localization.svg" class="img-fluid" alt="Seismic event localized from observations recorded at surface sensors"></p>
</div>
<p>The SEEM Group hosted Dr.&nbsp;Daniel Wamriew from the Seismic Wave Analysis Group at King Abdullah University of Science and Technology for a seminar titled <em>From Waveforms to Coordinates: Generative Localization of Microseismic Events</em>.</p>
<p>Dr.&nbsp;Wamriew presented work using generative diffusion models to estimate microseismic-event locations directly from waveform data. The approach offers a data-driven alternative to conventional localization workflows and is relevant to subsurface monitoring for CO₂ storage, hydraulic fracturing, and geothermal energy.</p>
<p>The visit concluded with discussions on potential collaboration between SEEM and KAUST in AI-driven, real-time seismic monitoring.</p>



 ]]></description>
  <category>Visitors</category>
  <category>Microseismic monitoring</category>
  <category>Generative AI</category>
  <guid>https://seem-kfupm.github.io/blog/posts/2025-11-19-kaust-generative-localization-seminar.html</guid>
  <pubDate>Wed, 19 Nov 2025 00:00:00 GMT</pubDate>
  <media:content url="https://seem-kfupm.github.io/assets/illustrations/seismic-localization.svg" medium="image" type="image/svg+xml"/>
</item>
</channel>
</rss>
