جستجوی استاد راهنمای دانشگاهی

Panagiota Birmpa is an Assistant Professor in the Department of Actuarial Mathematics & Statistics at Heriot-Watt University's School of Mathematical and Computer Sciences, Edinburgh. Her research bridges advanced mathematical theory with cutting-edge machine learning applications, focusing on uncertainty-aware methodologies for complex data systems. She actively supervises PhD students and maintains strong collaborative ties with international research groups in applied mathematics and computational science. Her academic credentials include: BSc+MSc (integrated master) in Applied Mathematics and Physical Sciences (majors in Analysis and Statistics) from National Technical University of Athens (NTUA), 2011 MSc in Pure Mathematics from National and Kapodistrian University of Athens (NKUA), 2014 PhD in Mathematics from University of Sussex, UK, 2018 (Thesis: Quantification of Mesoscopic and Macroscopic Fluctuations in Interacting Particle Systems) Dr. Birmpa's research program integrates theoretical mathematics with modern computational challenges through seven core domains: Generative modeling, Scientific Machine learning, Uncertainty Quantification, Probabilistic Graphical models, Interacting Particle Systems, Optimal transport Theory, and Partial Differential Equations. Her interdisciplinary approach enables innovative solutions for complex data analysis problems across scientific domains, particularly where traditional statistical methods face limitations in high-dimensional spaces. Analysis of her publication trajectory reveals a progression from foundational statistical physics (2017-2018 interface dynamics research) toward contemporary machine learning applications (2021-2024). Her recent work demonstrates increasing sophistication in merging deep learning architectures with uncertainty quantification frameworks, especially for scarce high-dimensional data scenarios where conventional approaches fail. This evolution reflects broader trends in mathematical data science toward robust, interpretable AI systems. No scientific awards or fellowships are currently listed in her professional profile. Dr. Birmpa accepts PhD candidates for projects exploring deep learning-graphical model interfaces with uncertainty quantification, building on her prior AFOSR-funded postdoctoral research at UMass Amherst (2021-2022). Her grant history includes significant support from the Air Force Office of Scientific Research for developing particle-based generative algorithms. She maintains active supervision of graduate researchers while pursuing methodological innovations in probabilistic modeling. Her collaborative research network spans multiple institutions including University of Massachusetts Amherst, with interdisciplinary teams developing novel mathematical frameworks for scientific machine learning. Current projects focus on Lipschitz-regularized gradient flows and generative particle algorithms for high-dimensional data, extending her earlier work on non-equilibrium fluctuations in particle systems.

• Generative modeling
• Scientific Machine learning
• Uncertainty Quantification
• +۴ مورد دیگر

Dongdong Chen serves as an Assistant Professor in the Department of Computer Science at Heriot-Watt University's School of Mathematical and Computer Sciences in Edinburgh, United Kingdom. His research profile indicates active supervision of PhD students with full scholarships available, demonstrating his established position within the academic community. Current research outputs show consistent publication activity from 2020 through 2024 with increasing impact. Dr. Chen's research interests center on machine learning applications for imaging systems, with particular expertise in image processing, computer vision, computational imaging, and inverse problems. His work bridges theoretical machine learning with practical applications, especially in medical imaging contexts. The fingerprint analysis of his publications reveals strong connections to Deep Learning (31%), Unsupervised Learning (25%), and Inverse Problems (43%), indicating a cohesive research trajectory focused on unsupervised frameworks for imaging challenges. Analysis of his 15 most recent publications (2021-2025) shows a clear evolution toward unsupervised and equivariant learning approaches for inverse problems, with increasing focus on diffusion models and medical applications. The research demonstrates strong theoretical foundations combined with practical implementations, particularly in MRI and medical diagnostics. Citation metrics indicate significant impact, with several papers exceeding 40 Scopus citations. His notable scientific achievements include: IES Best Student Paper Award (2014) MICCAI'18 BIA Best Paper Nomination (2018) Dr. Chen actively mentors PhD students and has attracted research attention through media coverage of his ICCV paper on Equivariant Imaging. His research contributes to UN Sustainable Development Goals through imaging technology development. Current activities include recruiting PhD students for projects in Edinburgh with full scholarship support, indicating ongoing research expansion and team development.

• Machine Learning
• Computer Vision
• Image Processing
• +۴ مورد دیگر

O. Deniz Akyildiz is an Assistant Professor in Statistics at the Department of Mathematics, Imperial College London. He holds a PhD in Signal Processing from Universidad Carlos III de Madrid (UC3M) and completed postdoctoral work at Warwick CS and The Alan Turing Institute. His research bridges computational statistics, machine learning, and applied probability, focusing on stochastic sampling methods, generative modeling, and optimization techniques for statistical inference and inverse problems. Education : BSc/MSc in Electronics and Communications Engineering (İTÜ), PhD in Signal Processing (UC3M). His work explores diffusion-based parameter estimation, score-energy-latent generative models, and overdamped/underdamped Langevin dynamics for probabilistic solutions to PDEs. He has developed adaptive importance samplers and contributed to high-dimensional stochastic filtering with real-world data. Recent efforts include lecture notes and code for stochastic simulation courses and organizing workshops on Gradient Flows for Sampling (Royal Statistical Society, 2023) and Accelerating Generative Models (Alan Turing Institute, 2022-2023). His publications emphasize error bounds in diffusion models, physics-informed neural processes, and convergence analysis of optimization-driven sampling algorithms.

• Computational Statistics
• Machine Learning
• Applied Probability
• +۷ مورد دیگر

Dr Andrew Valentine is currently an Associate Professor in the Department of Earth Sciences at Durham University, a position he has held since 2023. Previously, he served as Assistant Professor at Durham (2021-2023), Fellow at the Research School of Earth Sciences at The Australian National University (2016-2021), and Postdoctoral Researcher at Utrecht University (2011-2016). He holds significant leadership roles including Director of Education for the Department of Earth Sciences (2025-present) and Secretary of the IUGG Commission on Mathematical Geophysics (2023-present). DPhil in Earth Sciences, University of Oxford (2006-2010), supervised by Prof. J.H. Woodhouse BA/MSci in Natural Sciences (Physics), University of Cambridge (2002-2006) Valentine's research focuses on mathematical and statistical tools for extracting information from observational data, with expertise in geophysical inverse theory, global seismology, and machine learning applications in Earth Sciences. His work bridges theoretical mathematics with practical geophysical problems, developing novel approaches to data analysis and interpretation. He has made significant contributions to probabilistic inversion methods, seismic tomography, and the application of deep learning to geoscience problems. Analysis of his recent publications reveals a strong trend toward integrating machine learning with traditional geophysical methods. His work spans from theoretical developments in Bayesian inference and optimal transport to practical applications in seismic modeling, mantle dynamics, and volcanic processes. A notable pattern is his focus on uncertainty quantification in geophysical models and the development of efficient computational frameworks for complex inverse problems. His research increasingly incorporates neural networks and other AI techniques to address longstanding challenges in Earth Sciences. Harold Jeffreys Lectureship, Royal Astronomical Society (2025) ARC DECRA Fellowship (2018-2021) Multiple Geophysical Journal International Outstanding Reviewer citations Geophysical Journal International Student Author Award (2010) Valentine has supervised numerous PhD students including Buse Turunçtur, Matthias Scheiter, Suzanne Atkins, Paul Käufl, and Ralph de Wit, with thesis topics spanning sparsity-constrained inversion, Monte Carlo methods, mantle convection patterns, and probabilistic source inversion. His supervision extends to current students Arijit Chakraborty, Charlotte Aulton, Hanna-Riia Allas, Isaac Abbott, and Ugurcan Cetiner. He has served on various editorial boards including as Editor for Geophysical Journal International (2020-2025) and has been involved in major collaborative projects through organizations like CIG. As Director of InLab since 2018, Valentine leads a research group focused on developing innovative computational approaches to geophysical problems. His team has produced influential open-source software like pyprop8 for seismic modeling and has pioneered applications of deep learning to problems ranging from glacial isostatic adjustment to volcanic glass properties. The group maintains strong international collaborations, particularly with researchers at The Australian National University where Valentine holds an Honorary Senior Lecturer position (2021-2025).

• Geophysical inverse theory
• Global seismology
• Machine learning
• +۴ مورد دیگر

Professor István Z. Kiss is a Professor of Network Science at the Network Science Institute, Northeastern University London. His research bridges network science, dynamical systems, and stochastic processes with applications in epidemic modeling, computational neuroscience, and complex systems analysis. He leads the Systems Processes and Networks Lab (SPAN Lab) and maintains active collaborations across multiple institutions globally. Current position: Professor of Network Science Institution: Northeastern University London Research unit: Network Science Institute (SPAN Lab) Key collaborators: Péter L. Simon (Eötvös Loránd University), Joel C. Miller (La Trobe University), Gregory A. Rempała (Ohio State University) His research focuses on theoretical and data-driven problems at the intersection of network science and dynamical processes. Key areas include network inference, exactness of mean-field models, temporal and higher-order networks, adaptive/dynamic networks, and resilience of power networks. He has made significant contributions to understanding epidemic dynamics on complex networks, developing mathematical frameworks that connect approximate models with rigorous counterparts. Recent work emphasizes higher-order network structures, network inference from system-level data, and applications to public health and infrastructure resilience. His publication record demonstrates consistent output in top journals including Journal of Mathematical Biology, Bulletin of Mathematical Biology, and Physical Review E. The research shows a clear progression toward increasingly complex network structures, with recent publications focusing on higher-order interactions, temporal dynamics, and practical applications in epidemiology and infrastructure networks. His work bridges theoretical mathematics with real-world applications in disease control and network resilience. Professor Kiss actively supervises postdoctoral researchers and PhD students, with current advisees including Federico Cosimo Malizia (higher-order network contagion), Kevin Teo (shipping networks), and Yan Li (trade dynamics). His supervision spans mathematical theory, computational modeling, and data analysis across multiple domains. Leverhulme Trust grant RPG-2017-370: Bayesian Inference and Approximations in High-dimensional Network Models Network Science Institute Boston funding for higher-order network research EPSRC grant EP/H001085/1 for information diffusion modeling He leads the Systems Processes and Networks Lab (SPAN Lab), which develops mathematical frameworks for complex networked systems. The lab focuses on both theoretical advancements in network science and practical applications to epidemiology, neuroscience, and infrastructure networks. Current projects include developing tools for sequential temporal network analysis, understanding failure prediction in utility networks, and disentangling contact network structure effects on system-level outputs.

• Network Science
• Dynamical Systems
• Stochastic Processes
• +۶ مورد دیگر