Machine learning research and applications from the Gulf — spanning supervised learning, reinforcement learning, federated learning, and ML for scientific discovery.
This paper introduces Diffusion-BBO, a new online black-box optimization (BBO) framework that uses a conditional diffusion model as an inverse surrogate model. The framework employs an Uncertainty-aware Exploration (UaE) acquisition function to propose scores in the objective space for conditional sampling. The approach is shown theoretically to achieve a near-optimal solution and empirically outperforms existing online BBO baselines across 6 scientific discovery tasks.
KAUST researchers developed a new algorithm for detecting cause and effect in large datasets. The algorithm aims to find underlying models that generate data, helping uncover cause-and-effect dynamics. It could aid researchers across fields like cell biology and genetics by answering questions that typical machine learning cannot. Why it matters: This advancement could equip current machine learning methods with abilities to better deal with abstraction, inference, and concepts such as cause and effect.
MBZUAI researchers have developed a new kernel-based method to identify dependence patterns in data, especially in small regions exhibiting 'rare dependence' where relationships between variables differ. The method uses sample importance reweighting, assigning more importance to regions with rare dependence. Tested on synthetic and real-world data, the algorithm successfully identified relations between variables even with rare dependence, outperforming traditional methods like HSIC. Why it matters: This advancement can improve data analysis in fields like public health, economics, genomics, and AI, enabling more accurate insights from complex observational data.
Michael I. Jordan, a renowned AI researcher from UC Berkeley, has joined Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) as a laureate professor and honorary director of the Laureate Faculty Program. MBZUAI President Eric Xing highlighted Jordan's significant influence in machine learning, noting his role as a mentor. Jordan aims to guide AI researchers and advise the university as it seeks to become a global leader in AI. Why it matters: This appointment strengthens MBZUAI's position as a prominent AI research institution in the Middle East by attracting top-tier international talent and fostering a conducive environment for cutting-edge research.
A research paper by Dr. Reda Alami of the AI and Digital Science Research Center (AIDRC) at TII has been accepted for publication at the 14th Asian Conference on Machine Learning (ACML 2022). The paper addresses sequential decision-making under uncertainty in non-stationary environments, proposing a Bayesian Change-Point Detection with Thompson Sampling (Bayesian-CPD-TS) algorithm. The algorithm combines decision-making under uncertainty and sequential detection of abrupt changes. Why it matters: This recognition highlights the growing AI research capabilities within the UAE and its contribution to the global machine learning community.
McLaren Racing visited KAUST to celebrate their three-year R&D partnership, established in 2018, focusing on computational fluid dynamics, machine learning, and other areas. The visit included discussions on STEM education, sustainability, and a simulator race between McLaren driver Lando Norris and a KAUST community member. McLaren and KAUST have also collaborated on personal protective equipment, emergency ventilators, and sustainability initiatives. Why it matters: This partnership highlights the increasing role of advanced research and STEM education in Saudi Arabia's technology and sustainability initiatives, facilitated through collaborations between academic institutions and global industry leaders.
Researchers studied user lifetime prediction in the location-based social network Jodel within Saudi Arabia, leveraging its disjoint communities. Machine learning models, particularly Random Forest, were trained to predict user lifetime as a regression and classification problem. A single countrywide model generalizes well and performs similarly to community-specific models.
KAUST Associate Professor Xiangliang Zhang leads the Machine Intelligence and Knowledge Engineering (MINE) group, focusing on machine learning and data mining algorithms for AI applications. The MINE group researches complex graph data to profile nodes, predict links, detect computing communities, and understand their connections. Zhang's team also works on graph alignment and recommender systems. Why it matters: This research contributes to advancing machine learning techniques at a leading GCC institution, potentially impacting various AI applications in the region.
KAUST Professor Peter Richtárik received a Distinguished Speaker Award at the Sixth International Conference on Continuous Optimization (ICCOPT 2019) in Berlin. Richtárik's lecture series, totaling six hours, focused on stochastic gradient descent (SGD) methods, drawing from recent research by his KAUST group. He highlighted key principles and new variants of SGD, the key method for training modern machine learning models. Why it matters: This award recognizes KAUST's contribution to fundamental machine learning optimization, which is critical for advancing AI in the region.
A KAUST-led team in collaboration with Japan's National Institute of Informatics and Cray Inc. has implemented a new algorithm to harness the power of supercomputers. The algorithm integrates new singular value decomposition (SVD) codes into Cray LibSci scientific libraries, supporting machine learning and data de-noising applications. This was achieved through the Cray Center of Excellence (CCOE) at KAUST, established in 2015. Why it matters: The new algorithm helps to optimize the use of advanced supercomputing infrastructure in the region, specifically KAUST's Shaheen II, for computationally intensive AI applications.
KAUST Associate Professor Xiangliang Zhang presented her work on mining streaming and temporal data at the International Joint Conference on Artificial Intelligence and the European Conference on Artificial Intelligence (IJCAI-ECAI-18) in Stockholm. Her talk, "Mining Streaming and Temporal Data: from Representation to Knowledge," summarized her research on mining data streams. Zhang directs the KAUST Machine Intelligence and kNowledge Engineering (MINE) group, which focuses on knowledge discovery from large-scale data. Why it matters: Showcases KAUST's contributions to AI research and highlights the university's growing recognition within the international AI community.
KAUST and McLaren Racing have signed a five-year R&D agreement focused on extreme performance technology. The partnership will focus on computational fluid dynamics (CFD), machine learning, fuels and lubricants, advanced mathematics and sensors and electronics. The collaboration aims to advance research and offer talent development for KAUST graduate students through research and internships. Why it matters: This partnership highlights KAUST's growing role in international collaborations to advance research and development in key areas like AI and sustainable mobility solutions.
Ahmed Abdulmajeed Alabdulkarim, a KAUST alumnus (M.S. '11), pursued a Ph.D. at MIT and now leads a research lab at KACST and MIT. His research interests include big data, AI, and machine learning. He credits KAUST as a starting point for his growth as a scientist, providing a perfect research environment and interactions with distinguished scientists. Why it matters: The success of KAUST alumni in leading research roles at prominent institutions like KACST and MIT highlights the university's contribution to developing Saudi Arabia's AI research capacity.
Peter Richtárik, an associate professor of computer science and mathematics, joined KAUST in February 2017. He is affiliated with the Visual Computing Center and the Extreme Computing Research Center at KAUST. Richtárik's research combines optimization and machine learning, and he values the support KAUST provides to his students, including funding for travel and conference attendance. Why it matters: This highlights KAUST's commitment to attracting and supporting leading researchers in AI and related fields, fostering innovation and talent development in the region.
MBZUAI's inaugural class included eight Emirati master's graduates in computer vision and machine learning, making up 15% of the Class of 2022. All eight have secured employment or will pursue Ph.D. studies at MBZUAI, contributing to healthcare, technology, energy, transport, and government sectors in the UAE. Fatima Albreiki and Wafa Al Ghallabi will continue at MBZUAI for Ph.D. studies in Computer Vision. Why it matters: MBZUAI is nurturing local AI talent to support the UAE's national strategy and address global challenges in key sectors.
MBZUAI researchers presented a new machine learning method at ICLR for uncovering hidden variables from observed data. The method, called "complementary gains," combines two weak assumptions to provide identifiability guarantees. This approach aims to recover true latent variables reflecting real-world processes, while solving problems efficiently. Why it matters: The research advances disentangled representation learning by finding minimal assumptions necessary for identifiability, improving the applicability of AI models to real-world data.
MBZUAI researchers developed a new conditional independence test (DCT) that determines the dependence of two variables when both are discrete, continuous, or when one is discrete and the other is continuous. The new test addresses cases where variables are inherently continuous but represented in discretized form due to data collection limits. The findings will be presented at the 13th International Conference on Learning Representations (ICLR) in Singapore. Why it matters: This research addresses a fundamental problem in machine learning and statistics, improving causal relationship discovery in mixed datasets common across finance, public health, and other fields.
Tom M. Mitchell from Carnegie Mellon University discussed using machine learning to study how the brain processes natural language, using fMRI and MEG to record brain activity while reading text. The research explores neural encodings of word meaning, information flow during word comprehension, and how meanings of words combine in sentences and stories. He also touched on how understanding of the brain aligns with current AI approaches to NLP. Why it matters: This interdisciplinary research could bridge the gap between neuroscience and AI, potentially leading to more human-like NLP models.
Petar Stojanov from the Broad Institute of MIT and Harvard will give a talk on cancer data analysis, covering the fundamentals of cancer, the nature of large-scale data collected, and main analysis objectives. The talk will also address open questions in cancer data analysis and how machine learning and generative modeling can help. Stojanov's research focuses on applying machine learning to genomic analysis of cancer mutation and single-cell RNA sequencing data. Why it matters: Applying AI and machine learning to cancer research can lead to a better understanding of the disease and development of new therapies.
This article discusses a talk by Mingming Gong from the University of Melbourne at MBZUAI on bridging causality and machine learning. The talk focuses on using machine learning to discover causal structures from observational data, and leveraging causal structures to improve machine learning generalization and prediction in non-stationary environments. Gong's research explores theoretical foundations and computational innovations in causal structure learning from real-world data. Why it matters: This research direction is crucial for advancing AI systems that can reason about cause and effect, leading to more robust and reliable decision-making in complex environments.
A presentation will demonstrate the construction of well-calibrated, distribution-free neural Temporal Point Process (TPP) models from multiple event sequences using conformal prediction. The method builds a distribution-free joint prediction region for event arrival time and type with a finite-sample coverage guarantee. The refined method is based on the highest density regions, derived from the joint predictive density of event arrival time and type to address the challenge of creating a joint prediction region for a bivariate response that includes both continuous and discrete data types. Why it matters: This research from a KAUST postdoc improves uncertainty quantification in neural TPPs, which are crucial for modeling continuous-time event sequences, with applications in various fields, by providing more reliable prediction regions.
This talk discusses the asymptotic study of large asymmetric spiked tensor models. It explores connections between these models and equivalent random matrices constructed through contractions of the original tensor. Mohamed El Amine Seddik, currently a senior researcher at TII in Abu Dhabi, presented the work. Why it matters: The research provides theoretical foundations relevant to machine learning algorithms that leverage low-rank tensor structures, potentially impacting AI research and applications in the region.
MBZUAI's Eduardo da Veiga Beltrame is developing machine learning tools for analyzing single-cell RNA sequencing data, which measures RNA in thousands of individual cells. Sequencing costs have decreased faster than Moore's Law, enabling large-scale data collection in biology. RNA sequencing provides insights into gene expression and cellular activity, crucial for personalized medicine. Why it matters: Advancements in single-cell RNA sequencing and ML analysis will accelerate personalized medicine by providing detailed insights into cellular mechanisms and disease pathways.
MBZUAI researchers presented a new strategy for handling complex optimization problems in machine learning at ICLR 2024. The study, a collaboration with ISAM, combines zeroth-order methods with hard-thresholding to address specific settings in machine learning. This approach aims to improve convergence, ensuring algorithms reach quality solutions efficiently. Why it matters: Improving optimization techniques is crucial for advancing machine learning models used in various applications, potentially accelerating development and enhancing performance.
MBZUAI and KAUST researchers collaborated to present new optimization methods at ICML 2024 for composite and distributed machine learning settings. The study addresses challenges in training large models due to data size and computational power. Their work focuses on minimizing the "loss function" by adjusting internal trainable parameters, using techniques like gradient clipping. Why it matters: This research contributes to the ongoing advancement of machine learning optimization, crucial for improving the performance and efficiency of AI models in the region and globally.
MBZUAI PhD graduate William de Vazelhes is researching hard-thresholding algorithms to enable AI to work from smaller datasets. His work focuses on optimization algorithms that simplify data, making it easier to analyze and work with, useful for energy-saving and deploying AI models on low-memory devices. He demonstrated that his approach can obtain results similar to those of convex algorithms in many usual settings. Why it matters: This research could broaden AI accessibility by reducing computational costs, and has potential applications in sectors like finance, particularly for portfolio management under budgetary constraints.
MBZUAI Professor Kun Zhang is working on applying AI to understand cause-and-effect relationships in biology, with the goal of accelerating scientific discovery and improving human health. He aims to develop foundation models for biology that can process diverse data types and provide insights into the causes and treatments of health problems. These models could help scientists develop new medicines and preventative measures for diseases. Why it matters: This research has the potential to significantly advance the field of medicine by enabling a deeper understanding of the complex biological processes that underlie disease.
MBZUAI's Kun Zhang is applying causal machine learning to improve drug development and precision medicine, focusing on answering 'why' questions. Traditional drug development is costly (est. $2B) due to extensive studies needed to determine drug toxicity and efficacy. Zhang is combining causal ML with organs-on-chips technology to improve pre-clinical drug testing, aiming to reduce the failure rate of drugs in human trials. Why it matters: By improving the accuracy of pre-clinical drug testing, this research could significantly reduce the cost and time required to bring new medicines to market in the region and worldwide.
MBZUAI Professor Fakhri Karray and co-authors from the University of Waterloo have published "Elements of Dimensionality Reduction and Manifold Learning," a textbook on methods for extracting useful components from large datasets. The book addresses the challenge of the "curse of dimensionality," where growth in datasets complicates their use in machine learning. Karray developed the material from a popular course he taught at Waterloo. Why it matters: The textbook provides a unified resource for students and researchers in machine learning and AI, addressing a foundational challenge in processing high-dimensional data, relevant to diverse applications in the region.
MBZUAI held a pre-commencement celebration for its inaugural graduation on January 30, 2023, in Abu Dhabi. The first graduating class includes 52 students from 24 countries earning master's degrees in computer vision and machine learning. MBZUAI has quickly risen to become a top 25 institution globally in AI, with faculty from top 100 AI institutions. Why it matters: This milestone highlights the rapid growth and increasing importance of AI education and research in the UAE and the broader Middle East.
MBZUAI's Professor Le Song is developing an AI-driven simulation to model the human body at societal, organ, tissue, cellular, and molecular levels. The goal is to reduce the time and cost associated with bringing new medicines to market by removing the need for wet lab biological research. Song aims to create a comprehensive model using machine learning. Why it matters: This research could revolutionize drug discovery in the region by accelerating the development process and reducing reliance on traditional research methods.
Kevin Toner from Northern Ireland is the first student from NI to graduate from Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) in Abu Dhabi. He will receive a Masters in machine learning as part of the inaugural graduating class. Toner's interest in machine learning began during his computer science studies at Queen's University Belfast. Why it matters: This milestone highlights MBZUAI's growing international reach and its role in attracting global talent to the UAE's AI ecosystem.
MBZUAI Assistant Professor Samuel Horváth is researching federated learning to address the tension between data privacy and the predictive power of machine learning models. Federated learning trains models on decentralized data, keeping sensitive information on devices. Horváth's research focuses on designing algorithms that can efficiently train on distributed data while respecting user privacy. Why it matters: This work is crucial for advancing AI in sensitive domains like healthcare, where privacy regulations limit centralized data collection.
MBZUAI Professor Salman Khan is researching continuous, lifelong learning systems for computer vision, aiming to mimic human learning processes like curiosity and discovery. His work focuses on learning from limited data and adversarial robustness of deep neural networks. Khan, along with MBZUAI professors Fahad Khan and Rao Anwer, and partners from other universities, presented research at CVPR 2022. Why it matters: This research has the potential to significantly improve the ability of AI systems to understand and adapt to the real world, enabling more intelligent autonomous systems.
MBZUAI Professor Kun Zhang received a Test of Time Award Honorable Mention at ICML 2022 for his 2012 paper “On causal and anticausal learning." The paper, co-authored with researchers from the Max-Planck Institute, is considered foundational for causal learning in machine learning. Zhang's work demonstrated the importance of causality for machine learning tasks, helping to shift views in the field. Why it matters: This award highlights the growing recognition of causal AI research and MBZUAI's role in advancing the field.
MBZUAI has opened admissions for its M.Sc. and Ph.D. programs in computer vision and machine learning for the Fall 2021 semester. The university previously extended offers to 100 students for its first academic year commencing in January 2021, drawing from 2,223 applicants of 97 nationalities. MBZUAI has also completed its Masdar City campus, secured accreditation, appointed faculty, launched webinars, and formed partnerships with organizations like Virgin Hyperloop. Why it matters: As the first AI-focused graduate university in the UAE, MBZUAI is positioned to develop AI expertise and drive innovation aligned with the country's industrial and economic goals.
Dr. Pengtao Xie joins MBZUAI as an assistant professor focusing on healthcare and machine learning, inspired by human learning. He is developing automated machine learning methods for healthcare, such as neural architectures for pneumonia detection from chest X-rays. His method achieves state-of-the-art performance with 95% accuracy and is under review by Nature Scientific Report. Why it matters: This appointment strengthens MBZUAI's research capabilities in healthcare AI and signals the university's commitment to attracting top global talent to Abu Dhabi.
Dr. Kun Zhang from Carnegie Mellon University will spend 2022 as a Visiting Associate Professor in the Machine Learning Department at MBZUAI. Zhang's research focuses on causal discovery and causality-based learning, with applications in neuroscience, computer vision, computational finance, and climate analysis. He aims to develop methods for automated causal discovery from various kinds of data. Why it matters: This appointment strengthens MBZUAI's machine learning department and promotes research in causal AI, which is crucial for understanding and predicting complex systems.
MBZUAI Assistant Professor Bin Gu is working on black-box optimization techniques, especially in the context of vertical federated learning. Gu's work, in collaboration with JD.com, aims to enhance data and model privacy in machine learning. He is also focused on large-scale optimization and spiking neural networks to bring machine automation closer to the way the human brain operates. Why it matters: This research contributes to advancements in privacy-preserving machine learning techniques relevant to sensitive sectors like finance and healthcare in the region.
The Symposium on Data Mining and Applications (SDMA 2014) was organized by MEGDAM to foster collaboration among data mining and machine learning researchers in Saudi Arabia, GCC countries, and the Middle East. The symposium covered areas such as statistics, computational intelligence, pattern recognition, databases, Big Data Mining and visualization. Acceptance was based on originality, significance and quality of contribution.
Mladen Kolar from the University of Chicago Booth School of Business discussed stochastic optimization with equality constraints at MBZUAI. He presented a stochastic algorithm based on sequential quadratic programming (SQP) using a differentiable exact augmented Lagrangian. The algorithm adapts random stepsizes using a stochastic line search procedure, establishing global "almost sure" convergence. Why it matters: The presentation highlights MBZUAI's role in hosting discussions on advanced optimization techniques, fostering research and knowledge exchange in the field of machine learning.
Agathe Guilloux, a professor in Data Science at Evry Paris Saclay University, presented on machine learning algorithms for precision medicine at MBZUAI. Her talk covered the main challenges of precision medicine and how AI can address them. She also discussed algorithms developed for decision support tools. Why it matters: This highlights MBZUAI's role as a platform for discussing advanced AI applications in healthcare, even when the research is not directly conducted in the GCC.