Taiwo A. Adebiyi is a Ph.D. candidate in Civil Engineering at the University of Houston, where he conducts research in the Uncertainty Quantification Lab under Dr. Ruda Zhang. His work sits at the intersection of artificial intelligence, uncertainty quantification, Bayesian optimization, scientific computing, digital twins, and energy systems. At its core, his research asks a practical but difficult question: how can engineering systems make better decisions when data are limited, models are imperfect, and uncertainty cannot be ignored?
One of Taiwo’s defining doctoral contributions is TSRoots, an open-source algorithmic framework for efficient and exact Gaussian Process Thompson Sampling in Bayesian optimization. TSRoots advances autonomous experimentation by making it easier to choose the next most informative evaluation when simulations are expensive and design spaces are complex. The work was selected for an oral presentation at the NeurIPS 2024 Bayesian Decision-Making and Uncertainty Workshop in Vancouver, where it ranked among the top five percent of accepted papers and received Google DeepMind support. It later developed into a top eight percent paper at ICLR 2025 in Singapore, another premier global AI conference. Its growth from method development to open-source software and robust engineering design applications earned Taiwo the 2025–2026 Andrea Prosperetti Research Computing Student Award, one of the highest honors given by the UH Cullen College of Engineering in research computing.
As a UH Chevron Energy Graduate Fellow, Taiwo is applying probabilistic machine learning and Bayesian optimization to improve the reliability of geologic hydrogen resource estimation, an emerging clean-energy field where uncertainty is a major barrier to decision-making. This energy-focused trajectory will continue through his upcoming appointment as one of twelve 2026 Los Alamos National Laboratory Applied Machine Learning Fellows, where he will work on Bayesian machine learning for stochastic natural gas dynamics.
The various threads in Taiwo’s research culminate in his recent selection as one of five 2026 Chishiki AI Graduate Fellows through the NSF-supported Chishiki AI initiative. Through his selected project, “Trustworthy Adaptive Experimentation for Civil-Engineering Digital Twins,” he is expected to receive up to $49,000 in direct funding, along with travel support, computational resources, and mentorship to continue advancing trustworthy AI for engineering systems.
His Ph.D. journey has been marked by a growing sequence of competitive recognitions: the Cullen College travel award that supported his early digital-twin work; advanced computing training through the NHERI Computational Academy at the Texas Advanced Computing Center; Google DeepMind-supported recognition at NeurIPS; travel support to present at ICLR; the UH Chevron Energy Graduate Fellowship; the Andrea Prosperetti Award; the Los Alamos fellowship; and the NSF-supported Chishiki AI Fellowship. He was also part of a student team that won Best Technical Implementation at the Spring 2026 Coogs for Energy Hackathon. The team was later invited to present its energy-optimization decision-support product at Honeywell Automation Headquarters.
Taiwo serves as Vice President of the Civil Engineering Graduate Students Organization and holds leadership roles in his church community as well as reviews for leading research venues including NeurIPS, ICLR, and ICML, and mentors younger students. Before UH, he served as Campus Director for the United Nations Millennium Fellowship and led community-impact initiatives recognized through the Millennium Oceans Prize in Lagos, Nigeria.
Across research and service, Taiwo’s goal remains to build trustworthy computational intelligence that can move from theory into real value for engineering systems, energy resilience, and society.
Taiwo Adebiyi Talks About His Work at UH