KAUST Professor Håvard Rue was honored by the Royal Statistical Society (RSS) with the Guy Medal in Silver for his work on efficient computational techniques. The award recognizes Rue's contributions to the theory underpinning the INLA software, particularly through two influential papers on approximate Bayesian inference and Gaussian fields. Rue's research focuses on computational Bayesian statistics and Bayesian methodology, with the R-INLA project being a core part of his work. Why it matters: Recognition of KAUST faculty by international organizations highlights the institution's growing prominence in statistical research and computational modeling.
Professor Arnab Pain's group at KAUST discovered new insights on how a malaria protein enables parasites to spread malaria in human cells. Professor Haavard Rue's group upgraded the Integrated and Nested Laplace Approximation (INLA) for faster real-time modeling of large datasets. A KAUST-led study examined the stability of Y-series nonfullerene acceptors for organic solar cells. Why it matters: KAUST continues producing impactful research across diverse fields from medicine to climate change, advancing scientific knowledge and potential applications.
This paper introduces neural Bayes estimators for censored peaks-over-threshold models, enhancing computational efficiency in spatial extremal dependence modeling. The method uses data augmentation to encode censoring information in the neural network input, challenging traditional likelihood-based approaches. The estimators were applied to assess extreme particulate matter concentrations over Saudi Arabia, demonstrating efficacy in high-dimensional models. Why it matters: The research offers a computationally efficient alternative for environmental modeling and risk assessment in the region.
KAUST Ph.D. student Ghulam Qadir received a best poster award at the GRASPA 2019 conference in Italy. The winning poster, titled "Estimation of Spatial Deformation for Non-stationary Processes via Variogram Alignment," was based on Qadir's Ph.D. research project. The research focuses on developing covariance models for multivariate nonstationary random fields with applications to environmental data. Why it matters: This award recognizes KAUST's contribution to environmental statistics and highlights the university's commitment to advancing research in this area.
This article discusses methods for handling label noise in deep learning, including extracting confident examples and modeling label noise. Tongliang Liu from the University of Sydney presented these approaches. The talk aimed to provide participants with a basic understanding of learning with noisy labels. Why it matters: As AI models are increasingly trained on large, noisy datasets, techniques for robust learning become crucial for reliable real-world performance.
KAUST Professor Marc Genton received the 2024 Don Owen Award from the San Antonio Chapter of the American Statistical Association. The award recognizes Genton's excellence in research, statistical consultation, and service to the statistical community. Genton's research focuses on large-scale spatial and temporal data, with applications to environmental problems, including wind energy potential in Saudi Arabia. Why it matters: This award highlights KAUST's contributions to statistical research and its application to important environmental challenges in the region.
KAUST Professor Marc Genton has been selected as the 2020 Georges Matheron Lecturer of the International Association for Mathematical Geosciences. Genton will present a lecture at the 36th International Geological Congress in Delhi, India, focusing on geostatistics, climate model outputs, and the ExaGeoStat software developed at KAUST. His lecture will cover Matheron's theory of regionalized variables and showcase ExaGeoStat, a high-performance software for geostatistics with exascale computing capability developed at KAUST. Why it matters: This recognition highlights KAUST's contributions to advanced statistical methods and high-performance computing in geosciences, enhancing its international reputation in these fields.
Pascal Fua from EPFL presented an approach to implementing convolutional neural nets that output complex 3D surface meshes. The method overcomes limitations in converting implicit representations to explicit surface representations. Applications include single view reconstruction, physically-driven shape optimization, and bio-medical image segmentation. Why it matters: This research advances geometric deep learning by enabling end-to-end trainable models for 3D surface mesh generation, with potential impact on various applications in computer vision and biomedical imaging in the region.