The mm-wave 5G and beyond communication systems significantly improve the data rate, user capacity, and latency, however, the electromagnetic (EM) wave propagation suffers from high atmospheric attenuation as compared to the sub-6 GHz bands.

This presentation will focus on addressing the communication bottlenecks in distributed deep learning (DDL) training workloads. Deep neural networks (DNNs) are widely used in various domains, but training them can be time-consuming, especially with large models and datasets. Three innovative solutions are proposed and evaluated in the dissertation.

During this talk, we will delve into a new paradigm in modeling, called Hybrid Analysis and Modeling, which has the ability to combine the best of both the physics-driven and data-driven worlds, while eliminating their weaknesses. This approach has shown remarkable utility in the context of digital twin technology and will be demonstrated through a lab-scale experimental setup, mimicking building energy modeling.

Abstract

Recent advancements in inverse rendering have exhibited promising results for 3D representation, novel view synthesis, scene parameter reconstruction, and direct graphical asset generation and editing.

The dissertation focuses on developing novel computational methods to improve the diagnosis of patients with rare or complex diseases. By systematically relating human phenotypes resulting from gene function loss or change to gene functions and anatomical/cellular locations, the candidate aims to enhance the prediction and prioritization of disease-causing variants. These methods, leveraging graph-based machine learning and biomedical ontologies, demonstrate significant improvements over existing approaches. The presentation will include a systematic evaluation of the methods, demonstrating their ability to compensate for incomplete data and their applications in biomedicine and clinical decision-making. This research contributes to more effective methods for predicting disease-causing variants and advancing precision medicine, offering promising prospects for improved diagnostics and patient care.

Ontologies are widely used in various domains, including biomedical research, to structure information, represent knowledge, and analyze data. Combining ontologies from different domains is crucial for systematic data analysis and comparison of similar domains. This requires ontology composition, integration, and alignment, which involve creating new classes by reusing classes from different domains, aggregating types of ontologies within the same domain, and finding correspondences between ontologies within the same or similar domain.

Knowledge about transcription factor binding and regulation, target genes, cis-regulatory modules and topologically associating domains is not only defined by functional associations like biological processes or diseases, but also has a determinative genome location aspect.

Food loss and waste present a significant challenge to global sustainability, with approximately 1.3 million tonnes of food being lost or wasted each year. This not only leads to the depletion of resources but also contributes to greenhouse gas emissions. This dissertation focuses on the development and implementation of non-invasive solutions to extend the shelf life and monitor the quality of fresh foods, with the ultimate goal of reducing food loss and waste.

Power systems constitute a pillar of the critical infrastructure, and, as a result, their cybersecurity is paramount. Traditional power system architectures are moving from their original centralized nature to a distributed paradigm. This transition has been propelled by the rapid penetration of distributed energy resources (DERs) such as rooftop solar panels, battery storage, etc. However, with the introduction of new DER devices, the threat surface of power systems is inadvertently expanding.

On-site sensing systems provide fast and timely information about a myriad of applications ranging from chemical and biological to physical phenomena in the environment or the human body. Such systems are embedded in our daily life for detecting pollutants, monitoring health, and diagnosing diseases. This dissertation focuses on the design, development, and implementation of miniaturized PoC devices for achieving high sensitivity, selectivity, and reliability through a combination of hardware and software strategies at the edge.

We consider the incompressible axisymmetric Navier-Stokes equations as an idealized model of tornado-like flows. Assuming that an infinite vortex line that interacts with a boundary surface resembles the tornado core, we look for stationary self-similar solutions of the axisymmetric Euler and the axisymmetric Navier-Stokes equations emphasizing the connection among them as the viscosity ν → 0.

Wide bandgap III-Nitride semiconductor blue light-emitting diodes (LED) development has spawned the prestigious Nobel Prize in Physics in 2014. Building upon this success, the scope of research has expanded to ultrawide bandgap semiconductors, which possess immense potential in the realm of ultraviolet (UV) photonics. These materials have gained attention for their applicability in various areas, such as public sterilization, solar-blind UV communication, and real-time UV monitoring.

This thesis addresses the exponential growth of experimental data and the resulting computational complexity seen in two major scientific applications, which account for most cycles consumed on today's supercomputers.

Coffee Time: 15:30 - 16:00. Kernel matrices can be found in computational physics, chemistry, statistics, and machine learning. Fast algorithms for matrix-vector multiplication for kernel matrices have been developed, and is a subject of continuing interest, including here at KAUST. One also often needs fast algorithms to solve systems of equations involving large kernel matrices. Fast direct methods can sometimes be used, for example, when the physical problem is 2-dimensional. In this talk, we address preconditioning for the iterative solution of kernel matrix systems. The spectrum of a kernel matrix significantly depends on the parameters of the kernel function used to define the kernel matrix, e.g., a length scale.

“Practical fusion power is thirty years off and always will be.” Most people are instantly enraptured at the thought that a single glass of water will provide enough fusion fuel for their lifetime – if only it could be safely unlocked – and most of us have heard, as well, that we should not set our watches for when this will occur.

Generating fusion power from a stellarator is an exciting scientific and engineering challenge, one that promises to produce a stable, green, and nearly perennial energy source for mankind. To do so economically and safely will require a combination of several interdisciplinary breakthroughs including extreme scale computing (with adroit use of artificial intelligence and machine learning), high-temperature superconductors, materials science research, and robotics, to name a few. Overcoming this challenge will need a worldwide effort, involving academia, national laboratories, and industry, and contributions from diverse economies and people.

KAUST’s Extreme Computing Research Center and Program in Applied Physics are sponsoring an immersive Fusion Fest on the afternoon of June 5, 2023.  The program, which is designed to attract KAUST talent to address key challenges in bringing about commercially profitable “green” fusion power, consists of two live lectures by Professor Amitava Bhattacharjee and a hybrid live-virtual panel discussion (with Q&A) on a pathway to the realization of a commercial stellarator, led by physicist Bhattacharjee, with participation from engineering experts from ITER and the Princeton Plasma Physics Laboratory and a veteran financial advisor to the energy industry.

The most compelling transformational use of magnetically confined, high-temperature plasma is to realize sustained fusion energy. The tokamak, which is the leading magnetic confinement concept in the world today, first realized in 1958, has the geometry of a torus and toroidal symmetry, giving it good confinement properties. Nevertheless, the tokamak has a number of unresolved stability issues related to its current-carrying plasma that may be obstacles to its ultimate success.  In contrast, in the stellarator, the confining magnetic field is mostly produced by external current-carrying coils.

The Integrated Nested Laplace Approximations (INLA) method has become a commonly used tool for researchers and practitioners to perform approximate Bayesian inference for various fields of applications. It has become essential to incorporate more complex models and expand the method’s capabilities with more features. In this dissertation, we contribute to the INLA method in different aspects.

These discussions on the way how information and communication technology (ICT) should be in Beyond 5G (B5G) and 6G are accelerating, the space laser communication is becoming more advanced and active in the field of space communications.

The goal of this work is to investigate and advance a research on various topics, vital for the development of the future generations of optical communication technology. In the first part of the work, we present a fast and efficient simulation method of structured light free space optics (FSO) channel effects from propagation through the turbulent atmosphere.

It is widely acknowledged how the relentless surge of Volume, Velocity and Variety of data, as well as the simultaneous increase of computational resources have stimulated the development of data-driven methods with unprecedented flexibility and predictive power. However, not every environmental study entails a large data set: many applications ranging from astronomy or paleo-climatology have a high associated sampling cost and are instead constrained by physics-informed partial differential equations. Throughout the past few years, a new and powerful paradigm has emerged in the machine learning literature, merging data-driven and physics-informed problems, hence providing a unified framework for a whole spectrum of problems ranging from data-rich/context-poor to data-poor/context-rich. In this talk, I will present this new framework and discuss some of the most recent efforts to reformulate it as a stochastic model-based approach, thereby allowing calibrated uncertainty quantification.

The commonly used leave-one-out and K-fold cross-validation methods are not suitable for structured models with multiple prediction tasks. To overcome this limitation, we introduce leave-group-out cross-validation, which allows groups to adapt to different tasks. We propose an automatic group construction method and provide an efficient approximation for latent Gaussian models. Moreover, this method is conveniently implemented in the R-INLA software.

Latent Gaussian models (LGM) are widely used but struggle with certain datasets that contain non-Gaussian features, such as sudden jumps or spikes. This dissertation aims to provide tools for researchers to check the adequacy of the fitted LGM (criticism); if the check fails, offer efficient and user-friendly implementations of latent non-Gaussian models, which lead to more robust inferences (robustification).