The project supports the 54th Barrett Memorial Lectures at the University of Tennessee, Knoxville, centered on the mathematical foundations of data-driven scientific discovery. Many problems in science and engineering depend on extracting reliable information from high-dimensional, noisy, or limited data, requiring advances in mathematics and statistics to ensure models are accurate and trustworthy. This project brings together researchers, students, and early-career scientists to exchange ideas on methods with applications spanning biology, materials science, and engineering. Through lectures, poster sessions, and collaborative discussions, the project promotes interdisciplinary training, broad participation, and workforce development. By strengthening the mathematical foundations underlying artificial intelligence and data-driven modeling, the project contributes to national priorities in AI and biotechnology, and advances the national interest through scientific innovation, education, and the development of a diverse and skilled workforce. The project convenes researchers to investigate core mathematical and computational challenges in data-driven scientific modeling, with emphasis on uncertainty quantification, Bayesian inference, geometric and topological data analysis, and multiscale modeling. The Lectures integrate advances in probabilistic computation, Monte Carlo methods, numerical analysis, and scientific computing to address high-dimensional and data-scarce regimes. Particular focus is placed on unifying physical principles with machine learning through mathematically grounded frameworks that improve generalization, interpretability, and robustness. Activities include plenary and invited talks, poster presentations by junior participants, and structured breakout sessions targeting key research directions such as optimization, topological learning, and stochastic modeling. These interactions are designed to catalyze new collaborations and identify ope