Find Academic Supervisors

• Machine Learning
• Data Privacy
• Security
• +6 more

Adam Dziedzic is a Tenure Track Faculty Member at CISPA Helmholtz Center for Information Security , where he co-leads the SprintML group . His research focuses on Secure, Private, Robust, Interpretable, and Trustworthy Machine Learning , with applications to diffusion models, large language models, and collaborative learning. He previously held a Postdoctoral Fellow position at the Vector Institute and the University of Toronto , advised by Prof. Nicolas Papernot. He earned his PhD in computer science from the University of Chicago , advised by Prof. Sanjay Krishnan, and holds BSc and MSc degrees from Warsaw University of Technology in Poland. His work spans privacy attacks on image autoregressive models, dataset inference for large language models, differentially private collaborative learning, and defenses against model extraction. Recent publications (2025–2024) address privacy leakage in generative models, synthetic data calibration, secure noise sampling protocols, and memorization mechanisms in self-supervised and multi-modal architectures. Key contributions include: Developing POST for efficient soft prompt transfer in LLMs with privacy guarantees Proposing NeMo to localize and mitigate memorization in diffusion models Advancing Dataset Inference methods for diffusion and self-supervised models Designing secure multiparty computation protocols for collaborative learning His research leverages techniques like differential privacy , knowledge distillation , and statistical testing to balance utility and privacy across distributed systems.

Stefano FavaroView profile

Professor

• Nonparametric Bayesian Methods
• Empirical Bayes
• Statistical Machine Learning
• +5 more

Stefano Favaro is a Professor of Statistics at the University of Torino and holds the Carlo Alberto Chair of Statistics and Machine Learning at Collegio Carlo Alberto. His research focuses on Bayesian nonparametric methods, statistical machine learning, and applications to data confidentiality and deep learning theory. PhD in Statistics, Bocconi University, 2009 His research interests include: Nonparametric Bayes and Empirical Bayes methods Statistical Machine Learning with emphasis on learning-augmented algorithms Data confidentiality and fairness in machine learning Mathematical foundations of deep neural networks Recent work explores Bayesian nonparametric models for frequency estimation, deep learning asymptotics, and privacy-preserving statistical methods. He leads DataLab Algorithms at Collegio Carlo Alberto and serves as Associate Editor for Bernoulli and Statistical Science .

Björn OmmerView profile

Professor

• Computer Vision
• Machine Learning
• Deep Learning
• +7 more

Björn Ommer is a full Professor at Ludwig Maximilian University of Munich (LMU) where he heads the Computer Vision & Learning Group. Previously, he was a full professor at Heidelberg University and served as a director of the Interdisciplinary Center for Scientific Computing (IWR) and the Heidelberg Collaboratory for Image Processing (HCI). He is affiliated with multiple prestigious institutions including the Bavarian AI Council, ELLIS unit Munich, the Helmholtz Foundation, and the Munich Center for Machine Learning (MCML). Dr. Ommer received his PhD from ETH Zurich where he was awarded the ETH Medal for his dissertation 'Learning the Compositional Nature of Objects for Visual Recognition.' After completing his doctoral studies, he held a post-doctoral position in the Computer Vision Group of Jitendra Malik at UC Berkeley. His primary research interests span all aspects of semantic image and video understanding based on deep machine learning, with particular emphasis on generative approaches for visual synthesis (including Stable Diffusion), invertible deep models for explainable AI, deep metric and representation learning, and self-supervised learning paradigms. His work has significant interdisciplinary applications in digital humanities and neurosciences. His extensive publication record demonstrates a clear progression toward increasingly sophisticated generative models, culminating in the development of Stable Diffusion. His recent work focuses on improving diffusion models, exploring flow matching techniques, and developing more controllable generative systems with applications across multiple domains. German AI-Prize 2024 Technology-Prize of Eduard-Rhein-Foundation 2024 Nominated for German Future Prize of the President of Germany ELLIS Fellow ETH Medal for Dissertation Best Paper Award at CVPR'21 AI for Content Creation Workshop Professor Ommer serves as an associate editor for IEEE T-PAMI and has held significant leadership roles in major computer vision conferences including program chair for GCPR and Senior/Area Chair for CVPR, ICCV, ECCV, and NeurIPS. He delivered the opening keynote at NeurIPS'23 and has supervised numerous PhD students who have gone on to positions at leading technology companies including Amazon, Facebook, and Apple. His research group is located in downtown Munich and actively recruits talented students and researchers for cutting-edge work in computer vision and machine learning.

Martin ZachView profile

Research Fellow

• Biomedical Imaging
• Inverse Problems
• MRI Reconstruction
• +2 more

Martin Zach is a postdoctoral researcher at the Center for Biomedical Imaging (CIBM), EPFL , specializing in inverse problems in biomedical imaging. He joined the Mathematical Imaging Section under the supervision of Prof. Michael Unser in September 2024. PhD : Graz University of Technology (2024), advised by Thomas Pock Martin’s research bridges model-based reconstructions and data-driven approaches in imaging, with a focus on MRI reconstruction and diffusion models . His work explores regularization techniques, energy-based priors, and probabilistic modeling. Recent publications highlight advancements in Gaussian mixture models , non-linear inversion , and Langevin sampling for biomedical imaging. His Google Scholar profile reveals a strong focus on inverse problems (2020–2025), with applications in MRI , CT , and quantitative phase imaging . Key methodologies include diffusion models , generative priors , and regularization algorithms , spanning disciplines from machine learning to computational biology . Martin’s current role at EPFL involves collaborative research in mathematical imaging , with affiliations to the BioMedical Imaging (BIG) Group in Lausanne, Switzerland.

• Evolutionary Biology
• Population Genetics
• Ancient DNA Analysis
• +3 more

Professor Daniel Wegmann is affiliated with the Department of Biology at the University of Fribourg, within the Faculty of Science and Medicine. His research focuses on statistical and computational methods in evolutionary biology, particularly in analyzing genomic data from ancient DNA, population genetics, and conservation biology. His work bridges disciplines such as bioinformatics, ecology, and archaeology, addressing questions about human prehistory, species adaptation, and biodiversity conservation. Notable contributions include studies on Neolithic population dynamics, admixture patterns in ancient populations, and the application of Bayesian models to infer evolutionary processes. His group develops tools like ATLAS for low-depth sequencing analysis and contributes to open-source software in population genetics. Research interests emphasize the integration of genomic data with ecological and historical contexts, including projects on climate-driven range shifts, predator-prey dynamics in conservation, and the genomic origins of agricultural practices. Wegmann collaborates internationally, publishing in top journals like Science , Cell , and Nature Communications . His work often addresses applied challenges, such as quantifying ecological impacts and improving methods for analyzing sparse or degraded genetic samples. Key projects include investigating demographic shifts in post-Roman Europe through historic genomes, analyzing wolf-livestock conflicts in Europe, and exploring the genetic basis of ecotype formation in marine species. His lab’s activities highlight the interplay between theoretical modeling and empirical data, aiming to advance both fundamental and applied biological research.

Yawei LiView profile

Lecturer

• Artificial Intelligence
• Computer Vision
• Deep Learning
• +6 more

Yawei Li is a Lecturer at ETH Zürich, affiliated with the Integrated Systems Institute. His research focuses on efficient deep learning and AI systems for vision, language, and biosignals. Collaborating with Professors Luca Benini and Luc Van Gool, he explores topics like foundation models for biosignals, efficient neural network design, and deployment on RISC-V cores. He has organized workshops such as the NTIRE Challenge on Efficient Super-Resolution and served as a Senior Program Committee member for AAAI and IJCAI. He has supervised numerous PhD and Master students, with open positions available in foundational model development and bio-signal analysis. Key achievements include the Best Poster Presentation at ICVSS 2019 and contributions to projects like TinyYOLO for smart glasses. His work emphasizes practical applications of AI on embedded systems and low-power devices. Teaching includes courses like 'Deep Learning for Image Manipulation' and involvement in Computer Vision Lab initiatives. His research spans image restoration, neural network compression, and cross-modal analysis, with publications in top venues like CVPR, NeurIPS, and ECCV.

• International Economics
• Macroeconomics
• Economic Growth
• +4 more

Prof. Dr. Manuel Oechslin is a Full Professor of Economics at the University of Lucerne's Faculty of Economics and Management since 2014. Previously, he held an Associate Professorship at Tilburg University. His research focuses on international economics, macroeconomics, and the role of uncertainty in economic decision-making. He leads an SNSF-funded project exploring how fundamental uncertainty drives economic fluctuations and crises. Manuel Oechslin earned his PhD in Economics from the University of Zurich. His work has been published in top journals such as the Economic Journal , Journal of International Economics , and Journal of Economic Growth . His recent articles analyze geopolitical risks' impact on foreign investment (2025), behavioral macroeconomic models (2024), and transformative innovation under uncertainty (2023). These studies emphasize linkages between cognitive biases, institutional quality, and economic outcomes. While no specific awards are listed, his research has consistently addressed pressing issues like fiscal reforms, informal economies, and environmental scarcity. His work often intersects with policy design and institutional capacity-building in weakly institutionalized states. Grants and research projects include SNSF funding for his uncertainty-focused project. He advises doctoral candidates through the University of Lucerne's Graduate Academy and collaborates with interdisciplinary teams exploring topics like statistical governance and corruption dynamics.

Ilker OguzView profile

Lecturer

• Photonics
• Optical Computing
• Machine Learning
• +2 more

Ilker Oguz is Lecturer and Doctoral Assistant in the Laboratory of Applied Photonics Devices at EPFL. He teaches 'Advanced Photonics' and conducts research at the optics-AI interface. Research develops photonic computing architectures using nonlinear optics and multimode systems for machine learning acceleration. Key themes: optical neural networks, neuromorphic photonics, hybrid digital-optical systems. Publications demonstrate innovations in: Optical implementations of diffusion models Hardware-aware training paradigms Energy-efficient photonic processors Collaborates on European photonics initiatives and industry partnerships in computing hardware.

Niao HeView profile

Associate Professor

• Machine Learning
• Optimization
• Reinforcement Learning
• +4 more

Prof. Niao He is an Associate Professor in the Department of Computer Science at ETH Zürich. His research focuses on optimization theory, reinforcement learning, stochastic systems, and their applications in machine learning and multi-agent systems. He holds an academic position at one of the world's leading technical universities, contributing to both theoretical advancements and practical algorithmic solutions. His research interests span optimization theory (e.g., convex/non-convex optimization, stochastic optimization), reinforcement learning (policy gradient methods, multi-agent systems), and statistical learning (risk-averse methods, entropy regularization). He has also explored applications in data science, network revenue management, and deep reinforcement learning for complex systems. Recent work emphasizes algorithmic reproducibility, robustness under model uncertainty, and efficient methods for large-scale problems. Notable contributions include novel approaches to minimax optimization, mean-field games, and adaptive learning frameworks. His articles consistently address theoretical guarantees while maintaining practical relevance for real-world computational challenges.

Ender KonukogluView profile

Associate Professor

• Biomedical Imaging
• Computer Vision
• Medical Image Segmentation
• +3 more

Prof. Ender Konukoglu is an Associate Professor at the Department of Information Technology and Electrical Engineering at ETH Zürich, leading the Computer Vision Laboratory. He holds a PhD from the University of Nice Sophia Antipolis (2009), following B.S. and M.S. degrees from Bogazici University. His postdoctoral work included roles at Microsoft Research (2009–2012) and Harvard Medical School (2012–2016), where he contributed to computational neuroimaging and radiology. His research focuses on biomedical image computing, with emphasis on medical image segmentation, computer vision in surgery, and deep learning applications in healthcare. Key contributions include real-time anatomical guidance systems, multimodal image analysis, and predictive models for clinical outcomes. His work spans 3D reconstruction, radiomics, and domain adaptation in medical imaging. His research interests include leveraging computer vision and deep learning for medical diagnostics, surgical assistance, and quantitative imaging. Notable projects include the AENEAS initiative for endoscopic neurosurgery guidance and the development of benchmark datasets for cross-modality harmonization. He has published extensively on topics like camouflaged object segmentation, Gaussian splatting for scene understanding, and predictive analytics for cardiovascular procedures. Prof. Konukoglu’s awards and recognitions are not explicitly listed in the provided text. His team’s work on synthetic CT generation for radiotherapy and the Tera-MIND brain simulation project highlights his interdisciplinary impact. Ongoing efforts focus on robust medical image segmentation, domain generalization, and real-time surgical navigation systems.

• Financial Econometrics
• Quantitative Finance
• Risk Management
• +7 more

Olivier Scaillet is a Research Fellow at the Swiss Finance Institute, affiliated with the University of Geneva. His work spans financial econometrics, quantitative finance, and risk management, with a focus on stochastic volatility models, copulas, and high-frequency data analysis. Research Interests : Financial econometrics and nonparametric estimation Stochastic volatility and jump-diffusion models Asset pricing and factor models in large panels Systemic risk and recovery rate density estimation Machine learning applications in finance Market microstructure and high-frequency data dynamics Scientific Contributions : Developed methodologies for nonstandard error analysis in multi-analyst studies Advanced techniques for testing stochastic dominance efficiency and latent factor models Explored copula-based goodness-of-fit tests and threshold effects in time series Innovated in American option pricing under complex market conditions

• Financial Risk Management
• Extreme Value Theory
• Systemic Risk
• +3 more

Chen Zhou is a Professor at the Erasmus School of Economics, Erasmus University Rotterdam. His academic career is centered around quantitative finance and risk management, with significant contributions to extreme value theory applications in financial contexts. His research spans banking stability, market risk, and statistical methodologies for financial analysis. Chen Zhou's research interests focus on Financial Risk Management, Extreme Value Theory, Systemic Risk, Banking Stability, Financial Econometrics, and Asset Pricing. His work bridges theoretical statistical methods with practical financial applications, particularly in understanding tail risks and systemic vulnerabilities in financial markets. He has developed innovative approaches to measuring and managing financial risks using extreme value statistics, with applications across banking, equity markets, and energy markets. Analysis of Zhou's publication record reveals a strong focus on applying extreme value theory to financial risk measurement. His research shows a progression from foundational work in extreme value statistics to practical applications in banking stability and market risk. The papers demonstrate expertise in both theoretical statistical methods and their implementation in real-world financial contexts, with particular emphasis on systemic risk measurement and tail event analysis. Chen Zhou has collaborated extensively with researchers across Europe and globally, as evidenced by his co-authorship on the highly cited 'Nonstandard Errors' paper with over 50 co-authors. His research has been published in top finance journals including the Journal of Finance, and he has contributed to central bank research through working papers with institutions like De Nederlandsche Bank.

• Innovation Economics
• Causal Inference
• Quantitative Methods
• +4 more

Dr. Dragan Filimonovic is a Research Associate at the Center for International Economics and Business (CIEB), within the Faculty of Business and Economics at the University of Basel. He holds a PhD in economics from the University of Basel and specializes in causal inference and quantitative methods applied to innovation economics. His work focuses on the Swiss innovation landscape, including policy, productivity, and international collaboration. His research interests include Innovation Economics , Causal Inference , and Quantitative Methods , with a strong emphasis on the role of research, policy, and international integration in driving innovation. He investigates topics such as mRNA technology, intellectual property, and the impact of cross-border labor on innovation. Dragan's publications, primarily from 2020 to 2023, reflect a consistent focus on innovation policy, technological change, and the Swiss economy. His work often takes the form of policy reports and expert talks, featuring collaborations with leading economists and industry figures. The research spans fields such as biotechnology , productivity , pharmaceuticals , and science policy , demonstrating a multidisciplinary approach to understanding innovation systems. He is actively engaged in research dissemination through the CIEB's Innoscape series, contributing to public and academic dialogue on innovation. His collaborations include prominent researchers such as Rolf Weder, Matthias Niggli, and Christian Rutzer. Dragan Filimonovic does not have any listed scientific awards in the provided text. He does not appear to have formal advisees or students mentioned, but he contributes to team-based research projects and publications. He is associated with the Center for International Economics and Business , where he participates in research initiatives on European integration, trade, and innovation policy. The center serves as a hub for interdisciplinary research connecting economics with real-world policy challenges.

James HendersonView profile

Researcher

• Natural Language Processing
• Machine Learning
• Deep Learning
• +3 more

James Henderson is a Senior Researcher at Idiap Research Institute where he heads the Natural Language Understanding group. He currently serves as Action Editor for Transactions of the Association for Computational Linguistics (TACL) and was recently awarded an ERC Advanced Grant for his project 'Interpretable Beliefs and Programmable Knowledge with Bayesian Attention in Large Language Models' (BALM). Previously, Henderson held positions as Chargé de Cours at University of Geneva's Department of Computer Science and Principal Scientist at Xerox Research Centre Europe (now Naver Labs Europe). Henderson's research focuses on machine learning methods for natural language processing, with pioneering work on recurrent neural networks for syntactic and semantic parsing. His current investigations include representation learning for language semantics, graph-to-graph deep learning models, entity induction, and variational-Bayesian attention-based representation learning. His research bridges Bayesian inference, transformer architectures, and structured prediction for NLP tasks. His publication portfolio demonstrates consistent contributions to core NLP methodologies, with recent emphasis on transformer optimization, Bayesian neural methods, efficient model architectures, and graph-based language representations. Research frequently appears in top venues including ACL, EMNLP, ICLR, and NeurIPS. Honors: ERC Advanced Grant (2023) Henderson leads the Natural Language Understanding group at Idiap, currently recruiting PhD students and postdoctoral researchers for his ERC project. He obtained his PhD and MSc from University of Pennsylvania and BSc from Massachusetts Institute of Technology, all in computer science.

Taiji SuzukiView profile

Associate Professor

• Machine Learning
• Statistical Learning Theory
• Deep Learning
• +6 more

Prof. Taiji Suzuki is an Associate Professor at the University of Tokyo in the Department of Mathematical Informatics . He also serves as Team Leader of the "Deep Learning Theory" team at AIP-RIKEN , Japan. With a PhD in Information Science and Technology from the University of Tokyo (2009), he has held academic positions at the University of Tokyo (2009-2013) and Tokyo Institute of Technology (2013-2017) before returning to the University of Tokyo in 2017. University of Tokyo (2004: BEng in Mathematical Engineering) University of Tokyo (2006: MSc in Information Science and Technology) University of Tokyo (2009: PhD in Information Science and Technology) His research interests span statistical learning theory , deep learning , kernel methods , sparse estimation , and stochastic optimization . He investigates how neural networks adapt to function smoothness, avoid the curse of dimensionality, and achieve global optimization through mean-field dynamics. His work bridges theoretical guarantees (minimax optimality, convergence analysis) with practical implementations (transformers, graph neural networks, diffusion models). The 15 most recent articles focus on transformers' representation power, graph neural networks' limitations, diffusion models' convergence, and optimization theories for deep learning. Key themes include information-theoretic bounds , feature learning dynamics , and mean-field analysis . He explores applications in AI for medicine, federated learning, and biomedical modeling. Awards & Recognition: Outstanding Paper Award, ICLR 2021 MEXT Young Scientists’ Prize Outstanding Achievement Award, Japan Statistical Society 2017 Outstanding Achievement Award, Japan Society for Industrial and Applied Mathematics 2016 Best Paper Award, IBISML 2012 Best Paper Candidate, ICDM 2019 He has served as Area Chair for NeurIPS, ICML, ICLR, AISTATS, and as Program Chair for ACML. His scientific contributions include convergence theories for stochastic gradient methods, minimax analysis of deep learning vs kernel methods, and novel frameworks for distributional optimization in diffusion models.