Topics

<> proteins (3) <> Python (3) <> reliability testing (3)

Reproducibility in Computer Science - Part 1

-

B9 L3 R3223

Roy Maxion will give three lectures focusing broadly on different aspects of an increasingly important topic: reproducibility. Reproducibility tests the reliability of an experimental result and is one of the foundations of the entire scientific enterprise. We often hear that certain foods are good for you, and a few years later we learn that they're not. A series of results in cancer research was examined to see if they were reproducible. A startling number of them - 47 out of 53 - were not. Matters of reproducibility are now cropping up in computer science, and given the importance of computing in the world, it's essential that our own results are reproducible -- perhaps especially the ones based on complex models or data sets, and artificial intelligence or machine learning. This lecture series will expose attendees to several issues in ensuring reproducibility, with the goal of teaching students (and others) some of the crucial aspects of making their own science reproducible. Hint: it goes much farther than merely making your data available to the public.

Reproducibility in Computer Science - Part 2

-

B9 L3 R3223

Roy Maxion will give three lectures focusing broadly on different aspects of an increasingly important topic: reproducibility. Reproducibility tests the reliability of an experimental result and is one of the foundations of the entire scientific enterprise. We often hear that certain foods are good for you, and a few years later we learn that they're not. A series of results in cancer research was examined to see if they were reproducible. A startling number of them - 47 out of 53 - were not. Matters of reproducibility are now cropping up in computer science, and given the importance of computing in the world, it's essential that our own results are reproducible -- perhaps especially the ones based on complex models or data sets, and artificial intelligence or machine learning. This lecture series will expose attendees to several issues in ensuring reproducibility, with the goal of teaching students (and others) some of the crucial aspects of making their own science reproducible. Hint: it goes much farther than merely making your data available to the public.

Reproducibility in Computer Science - Part 3

-

B9 L3 R3223

Roy Maxion will give three lectures focusing broadly on different aspects of an increasingly important topic: reproducibility. Reproducibility tests the reliability of an experimental result and is one of the foundations of the entire scientific enterprise. We often hear that certain foods are good for you, and a few years later we learn that they're not. A series of results in cancer research was examined to see if they were reproducible. A startling number of them - 47 out of 53 - were not. Matters of reproducibility are now cropping up in computer science, and given the importance of computing in the world, it's essential that our own results are reproducible -- perhaps especially the ones based on complex models or data sets, and artificial intelligence or machine learning. This lecture series will expose attendees to several issues in ensuring reproducibility, with the goal of teaching students (and others) some of the crucial aspects of making their own science reproducible. Hint: it goes much farther than merely making your data available to the public.
<> reproducibility issues (3)

Reproducibility in Computer Science - Part 1

-

B9 L3 R3223

Roy Maxion will give three lectures focusing broadly on different aspects of an increasingly important topic: reproducibility. Reproducibility tests the reliability of an experimental result and is one of the foundations of the entire scientific enterprise. We often hear that certain foods are good for you, and a few years later we learn that they're not. A series of results in cancer research was examined to see if they were reproducible. A startling number of them - 47 out of 53 - were not. Matters of reproducibility are now cropping up in computer science, and given the importance of computing in the world, it's essential that our own results are reproducible -- perhaps especially the ones based on complex models or data sets, and artificial intelligence or machine learning. This lecture series will expose attendees to several issues in ensuring reproducibility, with the goal of teaching students (and others) some of the crucial aspects of making their own science reproducible. Hint: it goes much farther than merely making your data available to the public.

Reproducibility in Computer Science - Part 2

-

B9 L3 R3223

Roy Maxion will give three lectures focusing broadly on different aspects of an increasingly important topic: reproducibility. Reproducibility tests the reliability of an experimental result and is one of the foundations of the entire scientific enterprise. We often hear that certain foods are good for you, and a few years later we learn that they're not. A series of results in cancer research was examined to see if they were reproducible. A startling number of them - 47 out of 53 - were not. Matters of reproducibility are now cropping up in computer science, and given the importance of computing in the world, it's essential that our own results are reproducible -- perhaps especially the ones based on complex models or data sets, and artificial intelligence or machine learning. This lecture series will expose attendees to several issues in ensuring reproducibility, with the goal of teaching students (and others) some of the crucial aspects of making their own science reproducible. Hint: it goes much farther than merely making your data available to the public.

Reproducibility in Computer Science - Part 3

-

B9 L3 R3223

Roy Maxion will give three lectures focusing broadly on different aspects of an increasingly important topic: reproducibility. Reproducibility tests the reliability of an experimental result and is one of the foundations of the entire scientific enterprise. We often hear that certain foods are good for you, and a few years later we learn that they're not. A series of results in cancer research was examined to see if they were reproducible. A startling number of them - 47 out of 53 - were not. Matters of reproducibility are now cropping up in computer science, and given the importance of computing in the world, it's essential that our own results are reproducible -- perhaps especially the ones based on complex models or data sets, and artificial intelligence or machine learning. This lecture series will expose attendees to several issues in ensuring reproducibility, with the goal of teaching students (and others) some of the crucial aspects of making their own science reproducible. Hint: it goes much farther than merely making your data available to the public.
<> scientific enterprise (3)

Reproducibility in Computer Science - Part 1

-

B9 L3 R3223

Roy Maxion will give three lectures focusing broadly on different aspects of an increasingly important topic: reproducibility. Reproducibility tests the reliability of an experimental result and is one of the foundations of the entire scientific enterprise. We often hear that certain foods are good for you, and a few years later we learn that they're not. A series of results in cancer research was examined to see if they were reproducible. A startling number of them - 47 out of 53 - were not. Matters of reproducibility are now cropping up in computer science, and given the importance of computing in the world, it's essential that our own results are reproducible -- perhaps especially the ones based on complex models or data sets, and artificial intelligence or machine learning. This lecture series will expose attendees to several issues in ensuring reproducibility, with the goal of teaching students (and others) some of the crucial aspects of making their own science reproducible. Hint: it goes much farther than merely making your data available to the public.

Reproducibility in Computer Science - Part 2

-

B9 L3 R3223

Roy Maxion will give three lectures focusing broadly on different aspects of an increasingly important topic: reproducibility. Reproducibility tests the reliability of an experimental result and is one of the foundations of the entire scientific enterprise. We often hear that certain foods are good for you, and a few years later we learn that they're not. A series of results in cancer research was examined to see if they were reproducible. A startling number of them - 47 out of 53 - were not. Matters of reproducibility are now cropping up in computer science, and given the importance of computing in the world, it's essential that our own results are reproducible -- perhaps especially the ones based on complex models or data sets, and artificial intelligence or machine learning. This lecture series will expose attendees to several issues in ensuring reproducibility, with the goal of teaching students (and others) some of the crucial aspects of making their own science reproducible. Hint: it goes much farther than merely making your data available to the public.

Reproducibility in Computer Science - Part 3

-

B9 L3 R3223

Roy Maxion will give three lectures focusing broadly on different aspects of an increasingly important topic: reproducibility. Reproducibility tests the reliability of an experimental result and is one of the foundations of the entire scientific enterprise. We often hear that certain foods are good for you, and a few years later we learn that they're not. A series of results in cancer research was examined to see if they were reproducible. A startling number of them - 47 out of 53 - were not. Matters of reproducibility are now cropping up in computer science, and given the importance of computing in the world, it's essential that our own results are reproducible -- perhaps especially the ones based on complex models or data sets, and artificial intelligence or machine learning. This lecture series will expose attendees to several issues in ensuring reproducibility, with the goal of teaching students (and others) some of the crucial aspects of making their own science reproducible. Hint: it goes much farther than merely making your data available to the public.
<> solar energy (3)

Statistical Analysis of Multi-day Solar Irradiance using a Threshold Time Series model

-

B9 H2 R2325

Abstract The analysis of solar irradiance has important applications in predicting solar energy production from solar power plants. Although the sun provides every day more energy than we need, the variability caused by environmental conditions affects electricity production. Most of the existing statistical models to forecast solar irradiance are linear and highly depend on normality assumptions. However, solar irradiance shows strong non-linearity and is only measured during the day time. Thus, we propose a new multi-day threshold autoregressive (TAR) model to quantify the variability of the

Securing future energy supply

1 min read ·
KAUST’s Earth Science and Engineering Program is underpinned by a drive to develop sustainable ways to get the best from natural resources and quantify environmental hazards.

Measuring solar stores of the Arabian Peninsula

1 min read ·
Mapping variations in sunlight across the Arabian Peninsula reveals a bright future for solar energy in the region.
<> spatio-temporal data analysis (3)

Spatio-Temporal Data Analysis by Transformed Gaussian Processes

Yuan Yan, PhD Student of Prof. Marc Genton, KAUST

-

B2 L5 R5220

In the analysis of spatio-temporal data, statistical inference based on the assumption that the data follow a Gaussian distribution is ubiquitous due to its many attractive properties. However, data collected from different fields of science, such as meteorology, astronomy, biology and medicine, rarely meet the assumption of Gaussianity. One option is to apply a monotonic transformation to the data such that the transformed data have a distribution that is close to Gaussian. In this thesis, we focus on a flexible two-parameter family of transformations, the Tukey g-and-h (TGH) transformation

Ruiman Zhong

Ph.D. Student, Statistics
Ruiman is a Ph.D. candidate in Statistics at the King Abdullah University of Science and Technology (KAUST), studying under the supervision of Professor Paula Moraga in her Geospatial Statistics and Health Surveillance Research Group. Before joining KAUST, Ruiman obtained her Master's Degree in Statistics (August 2020) from Katholieke Universiteit Leuven, Belgium. Her research interests include spatial and spatial-temporal data modeling.

Marc G. Genton

Al-Khawarzmi Distinguished Professor, Statistics
Professor Genton’s pioneering research on large-scale spatial and temporal data has profoundly affected the field of environmental statistics. He is an expert in spatial and spatio-temporal statistics with applications to environmental problems.
<> Spectral Analysis (3)

A shock to the system

1 min read ·
The global, multifarious challenge posed by the COVID-19 pandemic has scientists tapping their wide-ranging fields of expertise to attack the problem on many fronts. Answering the call from KAUST President Tony Chan, and coordinated by the University's leadership team, KAUST researchers making up the Rapid Research Response Team (R3T) are turning this crisis into an opportunity to innovate.

Mara Sherlin Talento

Ph.D. Student, Statistics
We do not gain confidence through measuring what we know, but through quantifying how much we do not know. Mara Sherlin Talento is a PhD student in the Biostatistics group of King Abdullah University of Science and Technology, under the supervision of Prof. Hernando Ombao. She finished her Master’s degree in University of the Philippines and served the chairperson of Young Statisticians Committee of the International Statistical Institute during 2020 – 2023. She is also a member of American Statistical Association and International Biometric Society.

Hernando Ombao

Professor, Statistics
KAUST Professor of Statistics Hernando Ombao is an expert in time series models and their applications to designed medical experiments on complex biological processes.
<> statistical analysis (3)

Statistical Models and Methods Based on Stochastic Partial Differential Equations

-
This talk presents an overview of our research on statistical methods using stochastic partial differential equations (SPDEs), focusing on non-Gaussian random fields and fractional-order SPDEs, and theory for random fields and statistical analysis on metric graphs, highlighting theoretical contributions, software development, and applications relevant to KAUST RDI pillars.

Statistical Analysis of Multi-day Solar Irradiance using a Threshold Time Series model

-

B9 H2 R2325

Abstract The analysis of solar irradiance has important applications in predicting solar energy production from solar power plants. Although the sun provides every day more energy than we need, the variability caused by environmental conditions affects electricity production. Most of the existing statistical models to forecast solar irradiance are linear and highly depend on normality assumptions. However, solar irradiance shows strong non-linearity and is only measured during the day time. Thus, we propose a new multi-day threshold autoregressive (TAR) model to quantify the variability of the

Flexible Multivariate, Spatial, and Causal Models for Extremes

-

B4 L5 R5220

Risk assessment for natural hazards and financial extreme events requires the statistical analysis of extreme events, often beyond observed levels. The characterization and extrapolation of the probability of rare events rely on assumptions about the extremal dependence type and about the specific structure of statistical models. In this thesis, we develop models with flexible tail dependence structures, in order to provide a reliable estimation of tail characteristics and risk measures. Our novel methodologies are illustrated by a range of applications to financial, climatic, and health data.
<> statistical inference (3)

High-dimensional Q-learning for Dynamic Treatment Regimes

-

B3 L5 R5220

Dynamic treatment regimes are a set of decision rules and each treatment decision is tailored over time according to patients’ responses to previous treatments as well as covariate history. There is a growing interest in development of correct statistical inference for optimal dynamic treatment regimes to handle the challenges of nonregularity problems in the presence of nonrespondents who have zero-treatment effects, especially when the dimension of the tailoring variables is high.

Transparent and Robust Causal Inference in Data Science

Carlos Cinelli, Ph.D. candidate, Department of Statistics, UCLA

-

KAUST

The past few decades have witnessed rapid and unprecedented theoretical progress in the science of causal inference, ranging from the “credibility revolution” with the popularization of quasi-experimental designs, to the development of a complete solution to non-parametric identification with causal graphical models. Most of these theoretical progress, however, relies on strong, exact assumptions, such as the absence of unobserved common causes, or the absence of certain direct effects. Unfortunately, more often than not these assumptions are very hard to defend in practice. This leads to two undesirable consequences for applied quantitative work in the data-intensive sciences: (i) important research questions may be neglected, simply because they do not exactly match the requirements of current methods; or, (ii) researchers may succumb to making the required “identification assumptions” simply to justify the use of available methods, but not because these assumptions are truly believed (or understood). In this talk, I will discuss new theories, methods, and software for permitting causal inferences under more flexible and realistic settings. These tools empower scientists, and policymakers to both examine the sensitivity of causal inferences to violations of its underlying assumptions, and also to draw robust and trustworthy conclusions from settings in which traditional methods fail.

Alexandre de Bustamante Simas

Senior Research Scientist, Statistics
<> Survival analysis (3) <> wireless networks (3) <> 5G and beyond (2) <> Advanced Perception (2) <> AI4Science (2) <> air pollution (2) <> Applied Machine Learning (2) <> Artificial Neural Network (2) <> Automous Vehicles (2) <> biology (2) <> biomedical devices (2) <> cancer (2) <> climate data (2) <> climate projections (2) <> combustion (2) <> communication (2) <> communication networks (2) <> computational fluids mechanics (2) <> Cross-layer protocol design (2) <> data mining (2) <> drones (2) <> ecology (2) <> energy security (2) <> Energy-efficient wireless multihop networks (2) <> extreme environmental events (2) <> forecasting (2)

Statistical Analysis of Multi-day Solar Irradiance using a Threshold Time Series model

-

B9 H2 R2325

Abstract The analysis of solar irradiance has important applications in predicting solar energy production from solar power plants. Although the sun provides every day more energy than we need, the variability caused by environmental conditions affects electricity production. Most of the existing statistical models to forecast solar irradiance are linear and highly depend on normality assumptions. However, solar irradiance shows strong non-linearity and is only measured during the day time. Thus, we propose a new multi-day threshold autoregressive (TAR) model to quantify the variability of the

Cristian Felipe Jiménez Varón

Ph.D. Student, Statistics
Cristian Felipe Jiménez Varón is a Ph.D. candidate in the KAUST Environmental Statistics research group under the supervision of Professor Ying Sun. Before joining KAUST, Cristian obtained a master's degree in applied mathematics from the Universidad Nacional de Colombia Sede Manizales (UNAL), Colombia. He also holds dual bachelor's degrees in industrial engineering and chemical engineering from UNAL. Research Interests Cristian's research interests include the analysis of large volumes of data, statistical modeling and environmental statistics.
<> FPGA (2) <> Functional Time Series (2) <> Generative AI and LLMs (2) <> Geophysical Fluid Modeling (2) <> GPU Computing (2) <> graphical models (2) <> health (2) <> health surveillance (2) <> landslide susceptibility modeling (2)

Postdoc Luigi Lombardo gets an Outstanding Student Poster and PICO EGU Award. Congratulations Luigi!

1 min read ·
The 2016 Outstanding Student Poster and PICO (OSPP) Award is awarded to Luigi Lombardo for the poster/PICO entitled Landslide triggering-thickness susceptibility, a simple proxy for landslide hazard? A test in the Mili catchment (North-Eastern Sicily, Italy). Regarding his work, Luigi said: " In the PICO presentation project we modeled the thickness at the crown of each landslide as the dependent variable. As a result, the predictive maps carried more information than classic susceptibility ones, enabling a better proxy to hazard prediction." You can find Luigi's poster here.

Luigi Lombardo

Postdoctoral Research Fellow, Statistics
Dr. Luigi Lombardo joined the Extreme Statistics Research Group and worked as Postdoctoral Research Fellow from April, 2016, to December, 2018, under the joined supervision of Prof. Raphaël Huser and Prof. Martin Mai. After his postdoc at KAUST, Luigi moved to ITC, University of Twente, Netherlands, where he embraced an academic career by becoming an Assistant Professor of Soil Science. See his personal website here. Education and early career Luigi Lombardo received his Ph.D. from Universita’ degli Studi di Palermo, Palermo, Italy in 2015. He joined KAUST as a postdoc in April 2016, and moved
<> laser (2) <> latent Gaussian models (2) <> living systems (2) <> machine intelligence (2) <> machine learning for wireless communication (2) <> macroeconomics (2) <> Mathematical modeling (2) <> mathematical modelling (2) <> mathematics (2) <> mean-field games (2) <> multimodal sensing (2) <> Multivariate extremes (2) <> multivariate spatial processes (2) <> nanomaterials (2) <> Network Design (2) <> Neural Networks (2) <> Non-Terrestrial Networks (2) <> Nonlinear Optics (2) <> oceanography (2) <> optical computing (2) <> optimal control (2)

Epidemiological modeling and optimal control

-

KAUST

Compartmental epidemiological models are one of the simplest models for the spread of a disease. They are based on statistical models of interactions in large populations and can be effective in the appropriate circumstances. Their application historically and in the present pandemic has sometimes been successful and sometimes spectacularly wrong. In this talk I will review some of these models and their application. I will also discuss the behavior of the corresponding dynamical systems, and discuss how the theory of optimal control can be applied to them. I will describe some of the challenges in using such a theory to make decisions about public policy.

Raul Tempone

Professor Raul Tempone is an worlwide recognized researcher in numerical analysis and uncertainty quantification. Professor Raul Tempone at KAUST leads the stochastic numerics research group, significantly contributing to Saudi Arabia's Vision 2030 goals through advancements in computational science. His work in adaptive algorithms, Bayesian inverse problems, and scientific machine learning drives forward critical applications in technology and sustainability, embodying KAUST's commitment to global scientific leadership and economic diversification.
<> photonic integrated circuits (2) <> point processes (2) <> Programming language (2) <> quantile regression (2) <> Quantitative finance (2) <> quantum communication (2) <> R-INLA (2) <> remote sensing (2) <> renewable energy (2) <> risk assessment (2)

Flexible Multivariate, Spatial, and Causal Models for Extremes

-

B4 L5 R5220

Risk assessment for natural hazards and financial extreme events requires the statistical analysis of extreme events, often beyond observed levels. The characterization and extrapolation of the probability of rare events rely on assumptions about the extremal dependence type and about the specific structure of statistical models. In this thesis, we develop models with flexible tail dependence structures, in order to provide a reliable estimation of tail characteristics and risk measures. Our novel methodologies are illustrated by a range of applications to financial, climatic, and health data.

Yan Gong

Ph.D. Student, Statistics
Yan Gong was a Ph.D. student in Statistics at the King Abdullah University of Science and Technology (KAUST), under the supervision of Prof. Raphaël Huser. Yan successfully defended her PhD thesis entitled " Flexible Multivariate, Spatial, and Causal Models for Extremes" on March 28th, 2023; see her PhD thesis here. Her PhD committee was composed of Professors Raphaël Huser (chair), Valérie Chavez-Demoulin (external examiner from HEC Lausanne at UNIL, Switzerland), David Bolin, and Mohammed-Slim Alouini. For her next career steps, Yan has accepted a short-term postdoctoral research position at
<> robust statistics (2) <> scientific computing (2) <> simulation (2) <> Spatial and spatio-temporal statistics (2) <> spatio-temporal modeling (2)

CEMSE New Faculty: David Bolin, Associate Professor of Mathematical Statistics

1 min read ·
David Bolin is an associate professor of mathematical statistics who joined the KAUST Computer, Electrical and Mathematical Science and Engineering (CEMSE) Division earlier this month from the Department of Mathematical Sciences at the University of Gothenburg. Before joining KAUST, Bolin received both his Ph.D. degree in mathematical statistics and M.S.c in engineering mathematics from Lund University, Sweden, in 2012 and 2007, respectively. Upon completing his Ph.D., he spent one year at Umeå University, Sweden, working as a postdoctoral fellow before moving to the Chalmers University of Technology. In 2016, Bolin became an associate professor of mathematics at the University of Gothenburg, where a year later he received the title of Docent in mathematical statistics.

Pratik Nag

Ph.D. Student, Statistics
Pratik Nag is a Ph.D. candidate at the KAUST Environmental Statistics research group under the supervision of Professor Ying Sun. Before joining KAUST, Pratik obtained a master's degree in statistical quality control and operation research from the Indian Statistical Institute, India. Research Interests Pratik's research interests focus on machine learning, artificial intelligence and statistical techniques to solve problems related to spatio-temporal processes. Education Profile Master's degree in Statistical Quality Control and operation research, Indian Statistical Institute, India
<> SPDEs (2) <> Statistical computing (2) <> statistical signal processing (2) <> stochastic (2)

KAUST climate research named finalist for ADIA Lab Climate Data Sciences Award

3 min read ·
KAUST team develops breakthrough method for wind data generation, earning finalist spot for ADIA Lab’s climate science award.

Parameter Estimation for the McKean-Vlasov Stochastic Differential Equation

Nikolas Kantas, Associate Professor, Department of Mathematics, Imperial College London

-

KAUST

We consider the problem of parameter estimation for a McKean stochastic differential equation, and the associated system of weakly interacting particles. The problem is motivated by many applications in areas such as neuroscience, social sciences (opinion dynamics, cooperative behaviours), financial mathematics, statistical physics. We will first survey some model properties related to propagation of chaos and ergodicity and then move on to discuss the problem of parameter estimation both in offline and on-line settings. In the on-line case, we propose an online estimator, which evolves according to a continuous-time stochastic gradient descent algorithm on the asymptotic log-likelihood of the interacting particle system. The talk will present our convergence results and then show some numerical results for two examples, a linear mean field model and a stochastic opinion dynamics model. This is joint work with Louis Sharrock, Panos Parpas and Greg Pavliotis. Preprint: https://arxiv.org/abs/2106.13751
<> stochastic partial differential equations (2) <> STSDS (2) <> student focus (2) <> supercomputing (2) <> Time Series Analysis (2) <> UAV (2) <> ultrafast fiber lasers (2) <> Video Understanding (2) <> wind energy (2) <> wireless communication (2) <> xstat research highlights (2) <> 2D materials (1) <> 3D printing (1) <> 5G networks (1) <> 5th generation cellular (1)