Find Academic Supervisors

Dr. Andreas Nuessing is a researcher at the Institute for Analysis and Numerics within the Department of Mathematics and Computer Science at the University of Münster, Germany. His work focuses on numerical methods, particularly unfitted discontinuous Galerkin methods and algebraic multigrid techniques, applied to biomedical problems such as the EEG forward problem and brain stimulation (TDCS). Education: Dr. rer. nat. in Mathematics, supervised by Prof. Dr. Christian Engwer and Prof. Dr. rer. nat. Carsten Wolters. Research Areas include: Development of numerical methods for solving PDEs in brain research Efficient implementations of discontinuous Galerkin methods Applications in biomedical engineering and neuroscience Scientific Contributions : 2015: Best poster award at BACI2015 conference Key projects: Efficient solvers for DG discretizations of saddle point problems (2012–2013), German Academic Exchange Service collaborations

• Brain research
• EEG forward problem
• Brain stimulation (TDCS)
• +2 more

Dr. Henrik Schumacher is a researcher in the work group on Harmonic Analysis led by Prof. Dr. Philipp Reiter at TU Chemnitz. His research focuses on geometric variational problems, numerical optimization, and differential geometry, with applications in computer graphics and computational mathematics. He holds a PhD in Mathematics from TU Chemnitz (2015) and a Master's in Mathematics from the same institution (2010), where his thesis explored generalized Seiberg-Witten equations. His work bridges pure and applied mathematics, including contributions to self-avoiding energies, fractional Sobolev spaces, and algorithmic optimization for geometric problems. Notably, he co-developed the Repulsive Curves and Surfaces frameworks for collision-aware modeling, recognized by the SIGGRAPH 2024 Best Technical Paper Award . He also contributes to software tools like CoBarS and Repulsor for polygon sampling and energy optimization. Active in academic outreach, Schumacher supervises student computer tutorials on topics such as Euler elastica, finite element analysis, and geometric evolution equations. His teaching emphasizes computational methods and interdisciplinary applications, including contributions to TU Chemnitz's new Data Science programs. Awards: SIGGRAPH 2024 Best Technical Paper Award Software: CoBarS, Repulsor, ConformalBarycenter, PardisoLink Grants/Projects: Collaborations on applied analysis, geometric optimization, and inverse problems Lab/Team Affiliation: Harmonic Analysis Group at TU Chemnitz Faculty of Mathematics.

• Geometric Analysis
• Numerical Optimization
• Differential Geometry
• +3 more

Stephen L. Olivier is a prominent researcher in high-performance computing at Sandia National Laboratories, with a distinguished publication record spanning nearly two decades. His work focuses on parallel programming models, performance optimization, and energy-efficient computing across diverse architectures including CPUs, GPUs, and FPGAs. Olivier has made significant contributions to OpenMP standards and Kokkos programming model development, collaborating extensively with Department of Energy national laboratories and international research teams. Olivier's research interests center on task parallelism, memory management in distributed systems, and performance portability across heterogeneous architectures. His work addresses critical challenges in exascale computing, including efficient task scheduling for unbalanced workloads, power management in large-scale systems, and optimization of communication patterns. More recently, he has expanded his research into medical imaging applications, applying high-performance computing techniques to tuberculosis detection in rural healthcare settings. Analysis of Olivier's recent publications (2021-2024) reveals a strong focus on practical performance engineering for next-generation computing platforms. His work spans traditional HPC domains while increasingly incorporating data science applications and medical imaging analysis. The research demonstrates consistent innovation in parallel programming models, particularly around OpenMP tasking and Kokkos abstractions, with growing emphasis on energy efficiency and hardware-specific optimizations for emerging architectures. Olivier has maintained a prolific research output with numerous publications in top-tier conferences including SC, IPDPS, and IWOMP. His collaborative work extends across multiple Department of Energy laboratories and international institutions, reflecting the interdisciplinary nature of modern high-performance computing research. While specific grant information isn't detailed in the publication record, his work on DOE systems suggests significant involvement in national supercomputing initiatives.

• High-Performance Computing
• Parallel Programming
• Task Parallelism
• +6 more

Prof. Dr. Thomas Wick is a Professor at the Institute of Applied Mathematics within the Faculty of Mathematics and Physics at Leibniz University Hannover. He holds leadership roles, including Executive Director of the Institute and membership in the Executive Board and Faculty Council. His research focuses on numerical methods for coupled nonlinear partial differential equations, multiphysics systems (e.g., fluid-structure interaction, phase-field fracture), adaptive finite element techniques, and robust solvers. Key projects include the DFG-funded SPP 1962 and SPP 1748 initiatives, the PhoenixD Cluster of Excellence, and international collaborations like the Indo-German Higher Education Partnership. He has received grants from DFG, DAAD, and the Alexander von Humboldt Foundation. His work emphasizes algorithm design, error control, and computational efficiency in engineering and scientific applications. Research Interests: Numerical modeling of coupled PDE systems, multiphysics phenomena, phase-field fracture, adaptivity, and optimization. Projects include CoMeTeNd (IRTG 2657/1), PhoenixD Task Group S4, and Strukturerhaltende Adaptive Enriched Galerkin Methods. Scientific Awards: Feodor Lynen Fellowship, DFG Projects, DAAD grants. Collaborations span Germany, Austria, India, Peru, and France. Publications highlight advancements in phase-field fracture, fluid-structure interaction, and adaptive methods. His contributions address challenges in mesh adaptivity, error estimation, and high-performance computing.

• Numerical Analysis
• Computational Mechanics
• Partial Differential Equations
• +3 more

Dr. Sebastian Kuckuk is a researcher and head of training at the Erlangen National High Performance Computing Center (NHR@FAU), Friedrich-Alexander-Universität Erlangen-Nürnberg. He is affiliated with the Department of Computer Science and contributes to the Chair of System Simulation. His work bridges research, training, and software development in high-performance computing. Education: PhD in Computer Science, Friedrich-Alexander-Universität Erlangen-Nürnberg (2019) His research focuses on enhancing performance portability and programmer productivity using domain-specific languages, code generation, automatic parallelization, and GPU programming. These techniques are applied to develop massively parallel numerical solvers for computational fluid dynamics, particularly for the shallow water equations. He is a core developer of the ExaStencils framework, which enables automated generation of efficient multigrid solvers for structured and patch-structured grids. Analysis of his recent publications (2020–2025) reveals a consistent focus on code generation, GPU acceleration, and solver optimization for fluid dynamics problems. Key themes include heterogeneous computing, block-structured grids, and adaptive methods. His work integrates advanced compiler techniques with numerical mathematics to improve scalability and performance on modern HPC architectures. Scientific Recognition: NVIDIA Deep Learning Institute (DLI) University Ambassador Certified Instructor for DLI courses in GPU programming and CUDA He actively contributes to teaching and training through courses such as Programming Techniques for Supercomputers and High-End Simulation in Practice . He conducts workshops and tutorials on GPU programming and performance optimization. While no formal students are listed, his mentoring role is evident through collaborative research and training activities. He has no recorded grants in the provided text, but his involvement in NHR and KONWIHR projects indicates active participation in funded HPC initiatives. Laboratories and Projects: Lead developer of ExaStencils , a code generation framework for multigrid solvers Contributor to GHODDESS , a module for higher-order discretizations in shallow water modeling Active in NHR@FAU and KONWIHR projects focused on GPU computing and performance optimization

• High Performance Computing
• GPU Programming
• Parallel Computing
• +7 more

Robert Strzodka is a full professor at Heidelberg University, holding the chair on Application Specific Computing since 2015. He is renowned for pioneering work in GPU-based scientific computing, focusing on optimizing interactions between mathematics, algorithms, and parallel hardware architectures. Current affiliation: Institute of Computer Engineering (ZITI), Heidelberg University Prior roles: NVIDIA (AmgX group), Max Planck Institute (Integrative Scientific Computing), Stanford University (visiting assistant professor) PhD: Numerical mathematics from University of Duisburg-Essen (2004) His research spans parallel algorithms (GPUs, FPGAs), numerical methods (AMG, Krylov, preconditioners), and graph algorithms (partitioning, MST). Publications emphasize GPU acceleration , cache optimization , and mixed precision computing across domains like astrophysics, fluid dynamics, and finite element methods. The 15 most recent articles (2024-2008) demonstrate consistent innovation in parallel computing , GPU algorithms , and numerical methods . Key subfields include stochastic modeling , tridiagonal solvers , level set visualization , and cache-aware algorithms .

• Parallel algorithms on parallel hardware (GPUs, many-core CPUs, FPGAs)
• Numerical methods (AMG, Krylov, preconditioners, FMM, multilevel methods, SVD)
• Graph algorithms (partitioning, coloring, decomposition, BFS, MST)
• +7 more

Dr. Rene Simon is a researcher at the University of Duisburg-Essen , affiliated with the Faculty of Mathematics . His work focuses on nonlinear optimization and numerical methods for partial differential equations (PDEs). Education : Diploma in Technomathematics (2004) from TU Berlin; Doctorate (2008) from JKU Linz. Research interests include: Optimization with partial differential equations Finite element methods for PDEs Multigrid algorithms for control problems Numerical analysis of saddle point systems His publications highlight collaborations on SIAM Journal and Numerische Mathematik articles, emphasizing robust solvers and discretization techniques. He has taught courses ranging from Analysis to Mathematics for Engineers and Physicists.

• Nonlinear optimization
• Optimization with partial differential equations
• Finite element methods
• +3 more

Prof. Guillermo Gallego is a Professor of Electrical and Computer Engineering at Technische Universität Berlin since 2019. He holds a Ph.D. from Georgia Institute of Technology (2011), with postdoctoral research at Technical University of Madrid (2011–2014) and University of Zurich/ETH Zurich (2014–2019). His research focuses on computer vision, robotics, and ocean engineering, particularly in camera autocalibration, structure-from-motion, and variational methods for ocean wave reconstruction. He has contributed to stereo vision systems for oceanographic remote sensing, developing algorithms for 3D/4D wave reconstruction and calibration refinement. His work bridges geometric optimization, numerical methods, and PDEs, with applications in robotics and environmental monitoring. Education: Ph.D. in Electrical & Computer Engineering (Georgia Tech, 2011), Master's in Mathematics (Georgia Tech, 2009), Telecommunications Engineering (Technical University of Madrid, 2004) Key research interests include: Camera autocalibration via Plücker coordinates and line geometry Space-time reconstruction of ocean waves using variational stereo methods Efficient 3D reconstruction pipelines for uncalibrated systems His publications analyze statistical properties of ocean waves, stereo imaging constraints, and optimization techniques for dynamic scenes. Research has been applied to offshore platforms in the Black Sea and Adriatic Sea. He collaborates with oceanographic institutions to improve wave measurement technologies and extreme wave prediction.

• Computer Vision
• Robotics
• Signal Processing
• +7 more

Prof. Dr. Carsten Burstedde is a faculty member at the Institut für Numerische Simulation within the University of Bonn , specifically affiliated with the Faculty of Mathematics and Natural Sciences . His work focuses on developing scalable algorithms for adaptive mesh refinement (AMR) that operate efficiently on the world’s largest supercomputers. He leads the development of the p4est software library , a foundational tool for parallel AMR applications, and contributes to projects like ForestClaw for simulating volcanic ash transport in atmospheric flows. His research spans scientific computing , applied mathematics , and high-performance computing . Key application areas include geophysics (mantle convection, seismic wave propagation), fluid dynamics (incompressible flows, volcanic ash transport), and uncertainty quantification for inverse problems. He emphasizes non-conforming mesh techniques and parallel numerical solutions of PDEs , with a particular interest in hybrid mesh algorithms for complex domains. The 15 most recent publications highlight his expertise in adaptive mesh refinement across diverse contexts: 2021 works on heterogeneous systems and ghost layer optimization , 2020 contributions to p4est software and CPU ray tracing of AMR data, 2019 studies on Morton-type space-filling curves , and 2018–2016 projects enhancing ParFlow and ESPResSo with adaptive methods. Earlier papers (2015–2012) address Bayesian inverse problems , level-set methods , and multi-scale geodynamics . Scientific Awards include the Springer CSE Prize (2011) and the NSF TeraGrid Capability Computing Challenge (2008) . He has advised PhD student Johannes Holke , with whom he developed tetrahedral space-filling curves and hybrid AMR algorithms . His teaching includes courses on scientific computing , adaptive mesh refinement , and mathematics in music . The p4est summer school (2020) and collaboration with Donna Calhoun (ForestClaw project) underscore his leadership in computational science outreach and education.

• Scientific Computing
• Applied Mathematics
• Parallel Computing
• +3 more

Prof. Ulrich Rüde is a retired professor in the Department of Computer Science at Friedrich-Alexander-Universität Erlangen-Nürnberg, holding the Chair of Computer Science 10 (System Simulation). His research focuses on computational science, high-performance computing, lattice Boltzmann methods, and multigrid techniques. He has contributed to scalable simulation frameworks like waLBerla and HyTeG, emphasizing parallel algorithms and numerical methods for fluid dynamics and geodynamics. Editorial roles include SIAM Journal of Scientific Computing and International Journal of Parallel and Emergent Distributed Systems. Leadership in organizations: Zentralinstituts für Scientific Computing, Indo-German Winter Academy, and GAMM Computational Science and Engineering committee. Active in conferences: Copper Mountain Conference on Multigrid Methods, International Supercomputing Conference (ISC), and ACM Computing Frontiers. Research interests span fluid-structure interactions, particle-resolved simulations, and exascale computing. His work addresses challenges in parallel performance, algorithmic scalability, and efficient numerical solvers for complex systems.

• Computational Science
• High-Performance Computing
• Lattice Boltzmann Methods
• +4 more

Christian Schmitt is a researcher affiliated with the University of Erlangen-Nuremberg, Germany. He holds a PhD in Computer Science from the same institution (2019). His work focuses on high-performance computing, multigrid methods, domain-specific languages, and parallel algorithms. Schmitt has contributed to the development of ExaSlang and ExaStencils frameworks for scalable solver generation and stencil-code engineering. His research emphasizes automating solver development and optimizing performance on multi-core and FPGA platforms. Key research interests include parallel computing paradigms, reconfigurable hardware design, and domain-specific language technologies for scientific computing. His publications span journals like Proc. IEEE and conferences such as ASAP and FCCM. Collaborators include Frank Hannig, Jürgen Teich, and Harald Köstler. Recent work explores SYCL-based code generation for multigrid methods and OpenCL acceleration of discontinuous Galerkin methods. He has also investigated Scala prototypes for multi-platform solver implementations and stencil-code optimization using SPL Conqueror.

• Parallel Computing
• Multigrid Methods
• High-Performance Computing
• +3 more

Dr. Nils Kohl is a researcher in the Computational Science Department of Earth and Environmental Sciences at Ludwig-Maximilians-Universität (LMU) München. His work focuses on developing scalable numerical methods and high-performance computing frameworks for geophysical simulations, particularly in mantle convection and geodynamic modeling. He leads or contributes to projects like HyTeG, TerraNeo, and waLBerla, which aim to advance exascale computing and parallel algorithms for complex geophysical systems. His research emphasizes matrix-free finite element methods, multigrid solvers, and hybrid grid optimizations to address challenges in large-scale simulations. Kohl’s contributions include improving code generation for parallel computing, ensuring resiliency in extreme-scale simulations, and enhancing algorithmic efficiency for geodynamic applications. He maintains an active Google Scholar profile documenting his extensive publications on computational geophysics and high-performance frameworks. While no awards or student advisement details are explicitly stated, his involvement in TerraNeo and HyTeG underscores his role in advancing computational infrastructure for geoscience research. His work integrates interdisciplinary approaches to tackle grand challenges in earth system modeling.

• Geophysics
• Computational Science
• High-Performance Computing
• +3 more

Professor Dominik Göddeke is a Full Professor of Computational Mathematics at the University of Stuttgart, leading the Institute of Applied Analysis and Numerical Simulation. He concurrently serves as Vice Dean for Mathematics and co-spokesperson of the MaRDI consortium under the National Research Data Initiative. His expertise spans numerical methods for partial differential equations, high-performance computing, and hardware-oriented algorithm design. Education: PhD in Mathematics (2010) and Diplom in Computer Science (2004) from TU Dortmund. He has held academic positions at TU Dortmund (2011–2015) and the University of Stuttgart since 2015. His research focuses on scalable parallel algorithms, fault-tolerant computing, and interdisciplinary applications in geophysics and biomedical engineering. Research interests include GPU computing, domain decomposition methods, and green computing. His work emphasizes practical implementation techniques and software frameworks for large-scale simulations. Recent projects include the OpenDiHu framework for neuromuscular systems and preconditioners for Stokes-Darcy problems. His awards reflect teaching excellence and research impact, including the 2020 Digital Teaching Award and multiple best paper accolades. He actively contributes to international conferences and serves on editorial boards for computational science journals. Labs/Teams: Director of the Institute of Applied Analysis and Numerical Simulation, affiliated with the Stuttgart Centre for Simulation Sciences and the Cluster of Excellence for Data-Integrated Simulation Science.

• Numerical Analysis
• High Performance Computing
• Computational Science and Engineering
• +5 more

Martin Lanser is a Researcher at the University of Cologne's Department of Mathematics and Computer Science and a Core Scientist at the Center for Data and Simulation Science (CDS). His work focuses on developing efficient numerical methods for computational science and engineering problems, particularly targeting modern many-core architectures with million-way parallelism. His research spans Computational Science and Engineering, Numerical Methods, and Parallel Computing, with specialized expertise in Domain Decomposition Methods, Multigrid Approaches, and Computational Homogenization for heterogeneous solid mechanics. Lanser's theoretical work emphasizes nonlinear solvers for strongly heterogeneous materials, integrating algebraic multigrid techniques to enhance scalability in extreme-scale simulations. Analysis of his publication record reveals consistent innovation in scalable domain decomposition methods since 2014, with recent work targeting exascale computing through the FE2TI software framework. His research demonstrates strong interdisciplinary connections between computational mathematics, materials science, and high-performance computing, particularly in applications for dual-phase steel modeling. No scientific awards are documented in the provided information. Lanser's advising and grant activities are not specified in the source material, though his collaborative publications indicate extensive partnerships with researchers like Axel Klawonn and Oliver Rheinbach on projects including SCALEXA and High-Q club initiatives. As a core developer of the FE2TI software package—a computational homogenization implementation selected for the High-Q club—Lanser contributes to the CDS's research in quantitative modeling of complex physical systems. His work directly supports exascale computing projects focused on parallel domain decomposition methods and is integrated into the university's numerical analysis research infrastructure at numerik.uni-koeln.de.

• Computational Science and Engineering
• Numerical Methods
• Domain Decomposition Methods
• +4 more

Prof. Dr. Dietrich Braess is affiliated with the Ruhr University Bochum as a faculty member in the Faculty of Mathematics . His research focuses on numerical methods for elliptic differential equations, nonlinear approximation theory, and optimization, particularly the Braess paradox in traffic and network systems. Research Interests Multigrid methods for elliptic PDEs A posteriori error estimation via two-energies principle Finite elements in solid mechanics (locking phenomena) Stokes solvers and mixed elements Rational approximation and complex analysis methods Braess paradox in transportation and computer networks Co-Editorships Constructive Approximation (until 1999) SIAM Journal on Numerical Analysis (1991–2000) Numerical Linear Algebra European Transactions on Electrical Power (until 2010) Braess Paradox A counterintuitive phenomenon where adding roads can worsen traffic congestion. This has implications beyond transportation networks, extending to computer networks and electrical systems. Key papers include: Original 1968 German Paper 2005 English Translation

• Numerical Analysis
• Differential Equations
• Optimization
• +5 more