Find Academic Supervisors

Ivy PengView profile

Associate Professor

• High-Performance Computing (HPC)
• Cloud Computing
• Parallel Computing
• +6 more

Ivy Peng is an Associate Professor of Computer Science at KTH Royal Institute of Technology’s School of Electrical Engineering and Computer Science (EECS), where she leads the Scalable Computing Laboratory (ScaLab). She also holds a Docent title in Parallel Computing and is affiliated with the Digital Futures Faculty and the Cooperate Working Group. Prior to Sweden, she worked at the Lawrence Livermore National Laboratory (LLNL) as a Computer Scientist and as a post-doctoral fellow at Oak Ridge National Laboratory (ORNL) in the USA. Her research focuses on large-scale parallel systems, system-level optimization through workload and architecture awareness, performance modeling and analysis, and memory subsystem optimization. Key areas include converged HPC and cloud computing, heterogeneous accelerators (e.g., GPUs, RISC-V Vector units, QPUs, DPUs), and heterogeneous memories (e.g., persistent memory, HBM). She also explores disaggregated resources to enhance computing efficiency and scalability. Ivy Peng coordinates the EU Horizon 2022 Project OpenCUBE, which develops open-source cloud-based services for EPI systems, and leads the Swedish Research Council (VR) Grant on Disaggregated Memory for Emerging Parallel Systems. She actively contributes to technical committees and review boards for journals like IEEE TPDS and conferences including SC, ICS, HPDC, and IPDPS. She advises Ph.D. students Jacob Wahlgren, Gabin Schieffer, Ruimin Shi, and Daniel Medeiros, while co-supervising Jeremy Williams and Måns Andersson. Her academic leadership extends to multiple courses at KTH, including Applied GPU Programming, Introduction to High Performance Computing, and Quantum Computing for Computer Scientists.

Xuyang WuView profile

Associate Professor

• Distributed Optimization
• Machine Learning
• Federated Learning
• +3 more

Xuyang Wu is an Associate Professor at the School of Automation and Intelligent Manufacturing (AiM), Southern University of Science and Technology (SUSTech), China. Previously, he served as an Assistant Professor at SUSTech and held a postdoctoral position at KTH Royal Institute of Technology in Sweden under Professors Mikael Johansson and Sindri Magnússon. He completed his PhD in Communication and Information Systems at the University of Chinese Academy of Sciences (2015–2020) and a B.S. in Applied Mathematics from Northwestern Polytechnical University (2011–2015). His research focuses on distributed optimization, federated learning, and large-scale machine learning algorithms, emphasizing theoretical guarantees and practical implementations. Notable projects include safe resource allocation in power networks and asynchronous algorithms for distributed systems. He has published extensively in top-tier journals like IEEE Transactions on Automatic Control and conferences such as ICML and ICLR. Dr. Wu has received recognition including a best student paper finalist award at IEEE ICCA 2019. His work bridges theoretical advancements with real-world applications in smart networks and cyber-physical systems.

Jiaojiao ZhangView profile

Research Fellow

• Distributed Optimization
• Decentralized Machine Learning
• Algorithm Design
• +4 more

Jiaojiao Zhang is a Digital Futures Postdoctoral Fellow at KTH Royal Institute of Technology, focusing on decentralized machine learning algorithms with convergence guarantees. Her research addresses second-order, online, and robust methods for distributed optimization, emphasizing applications in privacy-preserving technologies and smart systems. Education: B.E. in Automation (Harbin Engineering University, 2015), M.Sc. in Control Theory (University of Science and Technology of China, 2018), Ph.D. in Operations Research (CUHK, 2022). Award: Hong Kong PhD Fellowship Scheme (2018). Supervisors: Mikael Johansson (KTH EECS) and Joakim Jaldén (KTH). Her work bridges distributed optimization and algorithm design, targeting challenges in federated learning, wireless sensor networks, and multi-core processor training. Projects like 'Decentralized Learning' aim to enhance cooperation and trust in digital futures through robust, efficient algorithms. Key areas: privacy-preserving federated learning, Byzantine-robust systems, and dynamic privacy allocation. Publications emphasize convergence analysis, communication efficiency, and adversarial resilience in decentralized settings. She actively contributes to Digital Futures, a joint initiative of KTH, Stockholm University, and RISE.

• High-Performance Computing (HPC)
• Cloud Computing
• Parallel Systems Optimization
• +5 more

Bo Peng is an Associate Professor of Computer Science at KTH Royal Institute of Technology in Stockholm, leading the Scalable Computing Laboratory (ScaLab). Previously, he held roles as a Computer Scientist at Lawrence Livermore National Laboratory (LLNL) and a postdoctoral fellow at Oak Ridge National Laboratory (ORNL). He coordinates the EU Horizon 2022 Project OpenCUBE and leads a Swedish Research Council (VR) grant on Disaggregated Memory for Emerging Parallel Systems. His research focuses on optimizing large-scale parallel systems through architecture-aware design, performance modeling, and memory subsystem innovation for heterogeneous computing environments. He teaches courses such as Applied GPU Programming, High-Performance Computing, and Operating Systems. As an academic leader, he supervises multiple PhD students and co-supervises additional researchers. His service includes roles on technical program committees for top conferences like IEEE TPDS, SC, and ICS, and journals like IEEE Transactions on Parallel and Distributed Systems. Bo Peng’s work has been recognized with prestigious awards, including the R&D 100 Award in 2024. His research bridges HPC and cloud computing, emphasizing scalable systems and resource efficiency in next-generation architectures.

Vincenzo GulisanoView profile

Associate Professor

• Distributed Systems
• Data Streaming
• Concurrent Data Structures
• +7 more

Vincenzo Gulisano is an Associate Professor in the Department of Computer Science and Engineering at Chalmers University of Technology. He holds a Ph.D. from the Technical University of Madrid and master's and bachelor's degrees from the University of Trieste. His research focuses on distributed systems, data streaming, and concurrent data structures with applications in cyber-physical systems such as Advanced Metering Infrastructures (AMI) and DDoS detection. He leads the Master Program in Computer Systems and Networks (MPCSN) and teaches courses like Operating Systems (EDA092/DIT400). His work emphasizes scalable and fault-tolerant stream processing frameworks like STRETCH and Erebus, which address challenges in real-time analytics and elastic resource management. Gulisano collaborates on projects integrating machine learning (e.g., IP-LSH-DBSCAN for parallel clustering) and cybersecurity (e.g., Metis for AMI intrusion detection). He has published widely in top-tier conferences like VLDB, Euro-Par, and IEEE TPDS, with contributions to stream synchronization, scheduling policies, and energy-sharing communities. Gulisano’s research bridges theoretical foundations and practical implementations, addressing scalability, determinism, and explainability in data-driven systems. His work often involves interdisciplinary collaborations, combining algorithms, hardware, and middleware design for efficient data processing in edge and cloud environments.

• Numerical Linear Algebra
• Data Assimilation
• Covariance Matrices
• +3 more

Jemima M. Tabeart is an Assistant Professor at TU Eindhoven, affiliated with the Centre for Analysis, Scientific Computing and Applications . Her research focuses on numerical linear algebra, data assimilation, and covariance matrices. She actively contributes to outreach and scientific crafting, and advises students on final projects in computational science. Education: PhD (2016–2019): University of Reading (supervised by Prof. Sarah Dance and others) MRes (2015–2016): Mathematics of Planet Earth CDT (Imperial College London/Reading) MMath (2011–2015): University of Bath with study abroad at Université Grenoble Alpes Research Interests: Large-scale numerical linear algebra and its applications Development of efficient preconditioners for 4D-Var and data assimilation systems Treatment of covariance matrices in computational models Interdisciplinary applications in environmental and meteorological modeling Teaching & Leadership: Organizing a 2025 Research Semester Programme at CWI on numerical analysis and machine learning Teaching courses on calculus, linear algebra, and optimization at TU Eindhoven Co-lecturing the MasterMath course on Advanced Numerical Linear Algebra Professional Activities: Member of the 4TU.AMI SRI in Bridging Numerical Analysis and Machine Learning Principal investigator in a network funded by the INI Network Support for Mathematical Sciences Advocate for low-carbon travel, documenting her journeys on her blog

August ErnstssonView profile

Lecturer

• Parallel Programming
• Heterogeneous Computing Systems
• High-Performance Computing
• +3 more

August Ernstsson is a Lecturer at Linköping University’s Department of Computer and Information Science (IDA). He is affiliated with the Graduate School in Computer Science (CUGS) and research groups Programming Environments (PELAB) and Software and Systems (SAS). His research focuses on parallel programming models, heterogeneous computing systems, and optimizing software for FPGA and GPU architectures. Notable contributions include frameworks for high development productivity (EXA2PRO) and deterministic parallel algorithms. Ernstsson’s work emphasizes performance portability, compiler design, and abstraction layers for heterogeneous systems. He collaborates with international teams on projects like pattern-based programming abstractions and FPGA-accelerated systems.

Prashant GoswamiView profile

Associate Professor

• Computer Graphics
• Virtual Reality
• Artificial Intelligence
• +7 more

Dr. Prashant Goswami is an Associate Professor in the Department of Computer Science at Blekinge Institute of Technology (BTH), Sweden, within the Faculty of Computing. He also serves as Adjunct Faculty in the CSE Department at IIIT Delhi, India. B.Tech/M.Tech (IIT Delhi) PhD in Computer Graphics (University of Zurich) His research focuses on Computer Graphics , Virtual Reality , AI Integration , and Real-Time Simulations , with recent work bridging Immersive Analytics and Human-Centered Intelligent Systems . He has contributed to snow/ice animation, foveated rendering, and XR ethics. Recent publications highlight trends in GPU-Based Simulation , GAN-Driven Cloud Animation , and AI-Enriched XR Systems . As Guest Editor for the Journal of Imaging (MDPI), he is advancing immersive metaverse experiences. Previously, he worked at gameLab (NTU Singapore) and INRIA's MAVERICK group, focusing on geometric modeling and cloud simulation. His industry experience preceded his PhD at University of Zurich.

Jose Nunez-YanezView profile

Professor

• Hardware Architectures
• Heterogeneous Systems
• Machine Learning Hardware
• +12 more

Jose Nunez-Yanez is a Professor in hardware architectures for Machine Learning at Linköping University, affiliated with the Department of Electrical Engineering and the Division of Electronics and Computer Engineering (ELDA). He has over 20 years of experience in high-performance embedded hardware design and is associated with the Wallenberg Autonomous Systems Program (WASP). His research focuses on energy-efficient, adaptive, and parallel hardware systems for machine learning and signal processing. Key areas include FPGA-based acceleration, run-time reconfiguration, and edge AI. He has led significant research projects with industry and academic partners such as ARM, ST Microelectronics, ESA, and Intel. The recent publications highlight a strong trend toward hardware-aware optimization of Graph Neural Networks for edge deployment, emphasizing quantization, dynamic range analysis, and heterogeneous computing. This reflects a broader shift toward efficient deep learning on resource-constrained platforms. Best Paper Award, Adaptive Hardware Systems (AHS) Conference, ESA/NASA, 2014 Best Paper Award, Adaptive Reconfigurable Computing (ARC) Conference, 2016 He has supervised 20 PhD students and served as the program director of the Advanced Microelectronics Systems Engineering (AMSE) program at the University of Bristol, significantly expanding its reach. He has secured substantial funding as Principal Investigator and Co-Investigator with partners including Sensata Systems, DSTL, Qioptiq, and Xilinx. His industry fellowships include Royal Society and Marie Curie awards. He is a member of the Royal Society Grants Board, the EPSRC College, and the Marie Curie Alumni Association. His work bridges academia and industry, with active consultancy in FPGA optimization and lossless data compression.

• Computer Science
• High Performance Computing
• Parallel Computing
• +3 more

Mikael Runnar is a Lecturer at the Department of Computing Science at Umeå University, actively contributing to academic and research activities. He is affiliated with the High Performance Computing Center North (HPC2N) , focusing on computational infrastructure development and optimization. Research Interests : His work spans High-performance computing Parallel processing Distributed system architectures Scientific application optimization Computational resource management Education in computer science . Affiliations : Umeå University (Department of Computing Science) HPC2N (High Performance Computing Center North)

Paolo BientinesiView profile

Professor

• High-Performance Computing
• Numerical Linear Algebra
• Tensor Operations
• +8 more

Paolo Bientinesi is a Professor at the Department of Computing Science, Umeå University, and Director of the High Performance Computing Center North (HPC2N). His research bridges theoretical and applied computer science, focusing on optimizing computational workflows through domain-specific innovations. Research Interests: Core Areas: Automatic generation of algorithms and code, numerical linear algebra, tensor operations, performance modeling, and computer music. Interdisciplinary Applications: Materials science, molecular dynamics, computational chemistry, computational biology, and computational physics. His work emphasizes leveraging architecture-specific and problem-specific knowledge to develop high-performance solutions. Publication Trends (2022-2026): Recent articles demonstrate a focus on mixed-precision computing, tensor decompositions, linear algebra algorithms (e.g., FLOPs optimization, matrix chains), and music information retrieval (e.g., automatic drum transcription, DJ cue points). Work frequently intersects with parallel computing, performance diagnostics, and machine learning. Leadership: Heads the research group High-Performance and Automatic Computing , driving projects in algorithm automation and computational efficiency.

Roman IakymchukView profile

Associate Professor

• Numerical Linear Algebra
• Parallel Programming
• High Performance Computing
• +3 more

Roman Iakymchuk is an Associate Professor at the Department of Computing Science within Umeå University, Sweden. His work focuses on numerically reliable and sustainable algorithmic solutions, emphasizing accuracy, reproducibility, and energy efficiency in high-performance computing environments. Research Interests: Numerical linear algebra, parallel programming models, sustainable computing, and optimization of distributed systems. Teaching: Courses on scientific computing, CUDA/OpenMP/MPI-based parallel programming, and floating-point error correction. The articles highlight his contributions to mixed-precision computing, numerical reliability in Krylov solvers, and energy-efficient practices in HPC systems. His work spans algorithm design, tool-assisted implementation, and European collaborative projects like CEEC.

• Coefficient inverse problems (CIPs)
• Adaptive finite-element methods
• Globally convergent numerical methods
• +5 more

Larisa Beilina is Full Professor at the Department of Applied Mathematics and Statistics, Chalmers University of Technology, Sweden. Her work sits at the intersection of numerical analysis, inverse problems, and computational electromagnetics, with recent extensions to biomedical modeling. Education and Career Milestones: PhD awarded in 2003 for pioneering an adaptive finite-element method for hyperbolic coefficient inverse problems. In 2008, co-developed—with Prof. M. V. Klibanov (UNC Charlotte)—a globally convergent numerical method for coefficient inverse problems. 2011: experimentally validated the globally convergent algorithm using blind field data from the US Army Research Laboratory’s Forward-Looking Radar. Research Interests Beilina’s research focuses on ill-posed and coefficient inverse problems governed by PDEs. She designs adaptive finite-element algorithms that deliver reliable reconstructions from limited, noisy data, and couples these with globally convergent optimization techniques to escape local minima. Application domains span microwave thermometry for hyperthermia monitoring , early-stage melanoma detection via RF/microwave imaging , subsurface imaging using ground-penetrating radar , and parameter identification in HIV dynamics models . Recent Publication Trends Over 2022-2024 her articles emphasize four themes: (1) stabilized and explicit finite-element schemes for Maxwell’s equations and coupled wave systems; (2) hybrid FEM/FDM domain-decomposition solvers for large-scale 3D electromagnetic inverse problems; (3) adaptive error control and a posteriori estimates ensuring quantitative accuracy in conductivity and permittivity reconstruction; and (4) biomedical applications —ranging from microwave melanoma detection to time-adaptive determination of drug efficacy in HIV infection models. Scientific Awards & Recognition Keynote speaker and session chair at multiple international conferences on inverse problems and computational electromagnetics. Serves on editorial boards and as guest editor for proceedings volumes such as Springer Proceedings in Mathematics & Statistics and Journal of Inverse and Ill-Posed Problems . Research Teams & Grants Leads a multidisciplinary group comprising graduate researchers and post-docs collaborating with partners including the US Army Research Laboratory, University of North Carolina at Charlotte, and medical institutions in Sweden. Funding sources encompass Swedish Research Council grants, EU Horizon programs, and industry contracts focused on microwave diagnostics and multiphase flow measurements. Laboratory & Infrastructure Heads the Computational Inverse Problems Laboratory at Chalmers, equipped with high-performance GPU clusters, anechoic chamber facilities for microwave experiments, and access to clinical data sets for melanoma and HIV projects.

• Computer Architecture
• High-Performance Computing
• Memory Systems
• +3 more

Per Stenstrom is a Professor of Computer Engineering at Chalmers University of Technology, Sweden since 1995. His research focuses on computer architecture, particularly high-performance memory systems and energy-efficient computing. He has authored/co-authored four textbooks, over 200 publications, 20 patents, and supervised approximately 25 PhD students. Awarded ACM Fellow and IEEE Fellow. Member of Academia Europaea and the Royal Swedish Academy of Engineering Sciences. Co-founder of the HiPEAC Network of Excellence. Served as Editor or Associate Editor for journals like ACM Transactions on Architecture and Code Optimization and IEEE Transactions on Computers . Program Chair for major conferences including ISCA, HPCA, IPDPS, and ACM International Conference on Supercomputing. His research innovations include cache compression techniques, memory optimization strategies, and energy-aware resource management in multicore systems. Ongoing work emphasizes secure and scalable cache partitioning, hybrid memory systems, and defense mechanisms against microarchitectural attacks.

• Numerical Analysis
• High Performance Parallel Computing
• Algorithm Design
• +5 more

Carl C Kjelgaard Mikkelsen is an Associate Professor at the Department of Computing Science, Umeå University, Sweden. His research focuses on developing robust numerical algorithms that cannot fail, particularly in the context of finite precision arithmetic used by computers. His work bridges theoretical numerical analysis with practical high-performance computing applications. Dr. Mikkelsen's research interests center around numerical analysis and high performance parallel computing, with special emphasis on the worst case analysis of algorithms. He investigates the profound differences between exact arithmetic used by humans and finite precision arithmetic used by computers. His current projects include developing parallel constraint solvers for constrained molecular dynamics and parallel solvers for ultra-sparse linear systems. He is a co-author of the StarNEig library for solving dense nonsymmetric standard and generalized eigenvalue problems in parallel. His recent publications demonstrate a strong focus on accuracy requirements in numerical methods, particularly Newton's method, and the development of robust parallel algorithms for linear algebra problems. His work spans from theoretical analysis of numerical stability to practical implementation in high-performance computing environments. Dr. Mikkelsen teaches numerical analysis and scientific computing, emphasizing algorithm design, analysis, and implementation. He focuses on teaching students the critical differences between exact and finite precision arithmetic, and invests significant time in teaching software development practices and technical writing.