Find Academic Supervisors

Sahand Jamal RahiView profile

Assistant Professor

• Biophysics
• Systems Biology
• Theoretical Physics
• +3 more

Sahand Jamal Rahi is an Assistant Professor at the École Polytechnique Fédérale de Lausanne (EPFL) in the School of Basic Sciences , leading the Laboratory of the Physics of Biological Systems since 2018. He has held academic appointments as a Visiting Scholar at Harvard University (2017-2018) and as an independent Fellow at The Rockefeller University (2010-2017). Education : PhD in Physics from MIT (2010), BA and MS from the University of Pennsylvania (2005), Vordiplom in Mathematics from Goethe-Universität Frankfurt (2003). His research focuses on the intersection of physics and biology , using dynamical systems to study biological circuit topologies , cell cycle control , and signal processing . He has published extensively on topics such as oscillatory stimuli , network topologies , and biochemical signaling , with recent work emphasizing the application of physical principles to biological systems. Key trends in his publications include analyzing frequency-detection properties of biological networks, elucidating cell cycle regulation through dynamical models, and resolving adaptive circuit topologies via oscillatory inputs. His work often bridges theoretical physics with experimental biology . Scientific Awards and Grants : NIH T32 Training Grant (2014-2017) Merck Fellowship (2013-2014) Fellowship at Rockefeller University’s Center for Studies in Physics and Biology (2010-2013) Lester Wolfe fellowship at MIT (2005-2006) Phi Beta Kappa (University of Pennsylvania) Vagelos Scholar in Molecular Life Sciences (University of Pennsylvania) His interdisciplinary approach has led to collaborations across institutions, including the Rockefeller University , Harvard Medical School , and MIT , with a focus on quantitative biology and biophysical modeling .

• Machine Learning
• Probabilistic Modeling
• Neural Networks
• +4 more

Carl Allen is a Laplace Junior Chair in Machine Learning at École Normale Supérieure, Paris, working in the research group of Stéphane Mallat, Giulio Biroli and Garbiele Peyré. Previously, he was a postdoctoral fellow at ETH Zurich and completed his PhD in Machine Learning in 2021 at the University of Edinburgh under the supervision of Professors Tim Hospedales and Iain Murray. His educational background includes a BSc in Mathematics & Chemistry from the University of Southampton, an MSc in Mathematics and the Foundations of Computer Science (MFoCS) from the University of Oxford, and MScs in Artificial Intelligence and Data Science from the University of Edinburgh. Before transitioning to AI/ML research, he spent several years in Project Finance. Allen's research focuses on mathematically understanding mechanisms behind successful machine learning methods, particularly neural networks. He investigates how machine learning models exploit aspects of data distribution from a probabilistic perspective. His current topics include explaining how VAEs disentangle independent factors of data, identifying mathematical models behind self-supervised learning, and deriving probabilistic interpretations of softmax classification. His PhD work investigated neural representations of discrete objects and their relationships, with a main result explaining how word embeddings can seemingly be added and subtracted (e.g., queen ≈ king - man + woman), which received Best Paper (honorable mention) at ICML 2019. His research spans theoretical foundations of machine learning, with particular emphasis on representation learning, disentanglement, and probabilistic modeling of neural networks. His work connects mathematical principles with practical machine learning applications, aiming to develop more interpretable and reliable algorithms. Allen has received notable recognition including a Best Paper honorable mention at ICML 2019 and a research grant from the Hasler Foundation. He has delivered invited talks at prestigious institutions including Harvard Center of Mathematical Sciences & Applications and Astra-Zeneca. His collaborative work spans multiple institutions, including a notable internship at Samsung AI Centre, Cambridge, where he worked at the intersection of representation learning and logical reasoning. His research has significant implications for developing more interpretable, reliable, and theoretically grounded machine learning systems.

Sagie BenaimView profile

Assistant Professor

• computer vision
• machine learning
• computer graphics
• +4 more

Sagie Benaim is an Assistant Professor at the School of Computer Science and Engineering at the Hebrew University of Jerusalem. Previously, he was a postdoctoral researcher at DIKU (Department of Computer Science, University of Copenhagen) working with Professor Serge Belongie and was a member of the Pioneer Center for AI. He completed his PhD at Tel Aviv University in the Deep Learning Lab under the supervision of Professor Lior Wolf. Dr. Benaim's research spans computer vision, machine learning, and computer graphics, with a particular emphasis on generative models, neural signal representations, and inverse graphics. His work explores how disentangled representations can be leveraged to better understand and manipulate visual content. He has made significant contributions to 3D scene manipulation, image-to-video generation, and neural rendering techniques. His recent publications reveal a strong trajectory toward advancing generative AI capabilities, particularly in 3D understanding, multimodal generation, and video synthesis. His research consistently bridges theoretical foundations with practical applications, demonstrating innovation in neural representation manipulation for both creative and analytical purposes. Dr. Benaim is actively seeking excellent students and postdocs to join his research group, indicating his commitment to mentoring the next generation of researchers in computer vision and machine learning. His position at the Hebrew University of Jerusalem places him within a vibrant academic community focused on advancing the frontiers of computer science.

• 3D Computer Vision
• Geometric Deep Learning
• Topological Deep Learning
• +10 more

Tolga Birdal is an Assistant Professor (Lecturer) and UKRI Future Leaders Fellow in the Department of Computing at Imperial College London. As the Principal Investigator (PI) of the CIRCLE group , his research focuses on topological deep learning, geometric machine learning, and 3D computer vision, with theoretical interests in non-Euclidean inference and deep learning principles. Education: PhD and MSc in Computer Vision from Technical University of Munich (2018), BSc in Computer Science from Sabancı University (2008). Projects: PI for UKRI-EPSRC's UNTOLD (Topological Deep Learning), Royal Society's drug discovery initiative, and EPSRC's GNOMON (Generative Models in non-Euclidean Spaces). Leadership: Area Chair for CVPR, ICCV, and 3DV 2025 Publication Chair. His work bridges differential geometry, algebraic topology, and deep neural networks, with applications in quantum computer vision, 3D/4D generative priors, and medical imaging. Key contributions include novel frameworks for rotation forecasting, graph generation, and topological generalization bounds. Scientific Awards: UKRI Future Leaders Fellowship EMVA Young Professional Award

Gavin BrownView profile

Professor

• Machine Learning
• Statistics
• Information Theory
• +3 more

Professor Gavin Brown is a faculty member in the Department of Computer Science, focusing on theoretical and methodological foundations of Machine Learning. His work bridges statistics, information theory, and information geometry to develop robust ML methods for reproducibility and hypothesis testing in non-standard scenarios. His research has been applied to predictive policing, clinical trials, and energy-efficient ML algorithms for plastic electronics. Key publications include A Unified Theory of Diversity in Ensemble Learning (JMLR, 2023) and On the Stability of Feature Selection in the Presence of Feature Correlations (ECML, 2019), supported by EPSRC funding. Recent work (2024) investigates the distinction between bias-variance and approximation-estimation error, alongside hardware-efficient ML solutions for FPGAs and flexible electronics. His book How to Get Your PhD (OUP, 2021) offers practical and emotional guidance for doctoral students, co-authored with 12 experts on career planning, diversity in science, and risk management in research. Developed theories for ensemble diversity (2020-2023) Created stability metrics for feature selection (2016-2022) Explored ML applications in healthcare and public safety Current activities include a research sabbatical (2021-2022) working on novel ML frameworks. He leads a research group at the Kilburn Building (Office G11) and contributes to pedagogy discussions, particularly around PhD training and conceptual understanding in deep learning.

Andrei BursucView profile

Lecturer

• Computer Vision
• Machine Learning
• Autonomous Driving
• +3 more

Andrei Bursuc is a Research Scientist and Deputy Scientific Director at Valeo.ai, where he focuses on perception for autonomous driving. Concurrently, he serves as a Research Associate at the Astra Inria project team and teaches the Deep Learning DIY course at Paris Institute of Technology (École Polytechnique) since 2019. He holds a PhD from Mines ParisTech and Telecom SudParis (2012) and a BSc from Polytechnic University of Bucharest (2009). His research spans computer vision and machine learning , with applications in autonomous driving, uncertainty quantification, self-supervised learning, and foundation models. Recent work emphasizes scalable representation learning, open-vocabulary segmentation, and generative modeling for driving scenarios. His publications (2024–2025) cluster around themes of semantic segmentation , depth estimation , and vision-language models , with innovations in unsupervised learning, cross-modal distillation, and efficiency. Trends indicate a focus on robustness, real-world generalization, and multi-sensor integration. He co-organizes workshops (e.g., Uncertainty Quantification at CVPR/ECCV) and serves as Area Chair for NeurIPS/CVPR/ECCV. He contributes to EU projects (e.g., ELLIOT) and open-source initiatives.

• Machine Learning in Biomedicine
• Statistical Genetics
• Systems Genetics
• +5 more

Dr. Francesco Paolo Casale serves as Principal Investigator in Machine Learning in Biomedicine at the Helmholtz Munich Institute AI for Health, part of Helmholtz Zentrum München and affiliated with Ludwig-Maximilians-Universität München's Biomedical Center. His research develops machine learning and statistical tools to analyze genetic cohorts with deep molecular and phenotypic data, addressing fundamental biomedical questions about disease mechanisms and progression. His academic foundation includes: PhD in Statistical Genetics from University of Cambridge & EMBL-EBI (2012-2016) M.Sc. in Physics of Complex Systems from Università di Napoli Federico II (2009-2012) B.Sc. in Physics from Università di Napoli Federico II (2009) Casale's research integrates machine learning, statistical inference, and systems genetics to develop scalable tools for genetic association studies, deep learning models for imaging genetics, and computational methods examining gene-environment interactions. His work emphasizes model robustness and interpretability while investigating molecular and cellular traits associated with disease severity. Current projects focus on rare variant analysis, aberrant gene expression prediction, and longitudinal omics data integration. His publication record shows a clear progression from foundational statistical genetics methods toward increasingly sophisticated integration of machine learning with multi-omics data. Recent work emphasizes practical biomedical applications including disease risk prediction through Mendelian randomization frameworks, advanced single-cell analysis techniques, and histopathology image classification. The research demonstrates consistent methodology development focused on scalability for large datasets while maintaining biological interpretability. Key recognitions include: Highly Recognized article in PloS Genetics Research Prize (2018) Microsoft Research New England Postdoctoral Fellowship (2017) EMBL studentship (2012) Honors for MSc and BSc degrees from Università di Napoli Federico II Throughout his career at Microsoft Research, Insitro, and Helmholtz Munich, Casale has led research teams developing computational approaches at the intersection of human genetics and machine learning. His work contributes to landmark projects including the 1000 Genomes Project and Blueprint initiative, with conference presentations at major venues including NeurIPS, ASHG, and EASL. Current grant support likely stems from Helmholtz Association funding mechanisms and collaborative biomedical research programs. He directs the Systems Genetics and Machine Learning Research team at Helmholtz Munich, which operates within the Biomedical Center ecosystem of LMU Munich. The team focuses on leveraging large-scale genetic datasets with machine learning to understand disease biology, with particular emphasis on target identification and characterization for therapeutic development. Current research directions include multi-timepoint omics analysis, disease subtyping, and developing interpretable models for clinical translation.

Bernhard EggerView profile

Assistant Professor

• 3D Morphable Models
• Statistical Shape Modeling
• Computer Graphics
• +7 more

Bernhard Egger is a junior professor (adidas Stiftungsprofessur) at the Chair of Visual Computing, Cognitive Computer Vision Lab at Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) . His research bridges human/machine perception of faces and shapes with synthetic data generation. Formerly, he was a postdoc at MIT's Computational Cognitive Science Lab and Computer Science & AI Lab, following a PhD in facial image annotation at the University of Basel. Research Focus: 3D Morphable Models, Statistical Shape Modeling, Inverse Rendering, AI in Mental Health, Medical Imaging. Education: PhD (University of Basel, 2017), MSc/BSc in Computer Science (University of Basel), Teaching Diploma (University of Applied Sciences Northwestern Switzerland). Notable Awards: Best Poster Award (2024, Cognitive Computational Neuroscience) Best Paper Award (CVPR Workshop 2019) Honk Award (SIGBOVIK 2020) Publications Trends: Recent work spans implicit surface modeling (ICLR 2025), 3D scene decomposition (3DV 2025), medical shape models (BVM 2025), and multimodal AI (npj Mental Health 2024). Labs: FAU's Cognitive Computer Vision Lab (advisor to 8+ students/researchers).

• Computer Vision
• Egocentric Vision
• Human-Computer Interaction
• +7 more

Antonino Furnari is a tenure-track Assistant Professor at the University of Catania's Department of Mathematics and Computer Science, where he is a member of the Image Processing Laboratory (IPLAB). His research focuses on egocentric vision, investigating how intelligent systems can perceive, understand, and anticipate human actions and interactions from an embodied, first-person viewpoint to enable assistive technologies on wearable devices. Dr. Furnari teaches a Bachelor course on Computer Programming and a Master course on Fundamentals of Data Analysis at the University of Catania. He has supervised more than 30 bachelor and master theses and is currently supervising 8 PhD students. His research has resulted in numerous publications in top-tier computer vision and artificial intelligence conferences and journals. His research interests span multiple areas within computer vision and artificial intelligence, with a strong focus on egocentric vision , human action recognition , procedural understanding , and assistive technologies . His work often involves developing novel deep learning approaches for analyzing first-person video data, with applications in industrial settings, kitchen environments, and cultural sites. He has been instrumental in creating important datasets in the field, including EPIC-KITCHENS and MECCANO. Dr. Furnari serves in prestigious editorial roles, including as an Associate Editor for IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) and as an Area Chair for WACV 2026. He also served as publicity co-chair for ICCV 2025, one of the top conferences in computer vision. His research has been widely recognized in the computer vision community, with numerous publications at top venues including CVPR, ECCV, NeurIPS, and IJCV. His work focuses on advancing the state-of-the-art in egocentric vision, with particular emphasis on action anticipation, procedural understanding, and developing practical applications for assistive technologies.

• Signal Processing
• Machine Learning
• Tracking Algorithms
• +3 more

Ángel García-Fernández is an Associate Professor at the Polytechnic University of Madrid , specializing in Bayesian inference , multi-target tracking , and nonlinear filtering with applications in signal processing, robotics, and underwater mapping. His work includes the development of Poisson multi-Bernoulli mixture (PMBM) filters, iterated posterior linearization algorithms, and direction-of-arrival (DOA) measurement models.

• Probabilistic machine learning
• Bayesian experimental design
• Causal inference
• +4 more

Desi R. Ivanova is a research fellow at the University of Oxford's Department of Statistics under the Florence Nightingale Bicentennial Fellowship. Her work bridges probabilistic machine learning, Bayesian experimental design, and LLM evaluation frameworks. She holds a DPhil in Statistics from Oxford's StatML CDT program (2020-2024) and an MMORSE in Mathematics from University of Warwick (2011-2016) with Erasmus exchange at LMU Munich. Research spans causal machine learning and uncertainty quantification Developed CO-BED and Step-DAD frameworks Focus on LLM evaluation methodology and calibration Expert in real-time adaptive experimental systems Her publications demonstrate expertise in Bayesian self-consistency methods, neural data compression, and privacy-preserving dataset merging. Key contributions include improving amortized inference efficiency and developing gradient-based causal experimental designs. Current work emphasizes rigorous statistical evaluation of language models, advocating for appropriate uncertainty quantification when analyzing performance across small datasets. She critiques CLT-based methods for LLM evaluation and proposes more robust frequentist and Bayesian alternatives.

Melih KandemirView profile

Associate Professor

• Machine Learning
• Bayesian Inference
• Stochastic Process Modeling
• +5 more

Melih Kandemir is an Associate Professor of Machine Learning at the University of Southern Denmark, Department of Mathematics and Computer Science. He earned his PhD in 2013 from Aalto University under Prof. Samuel Kaski, followed by postdoctoral work at Heidelberg University (with Prof. Fred Hamprecht) and an assistant professorship at Ozyegin University, Turkey. Prior to joining SDU, he led research at Bosch Center for Artificial Intelligence. Education: PhD (Aalto University, 2013), Postdoc (Heidelberg University). Previous Roles: Assistant Professor (Ozyegin University), Research Group Leader (Bosch CAI). His research focuses on Bayesian inference, stochastic process modeling with deep neural networks, and applications to reinforcement learning and continual learning. He leads the SDU Adaptive Intelligence (ADIN) Lab and is an ELLIS Member, reflecting his standing as a top European AI researcher. His work addresses critical challenges in uncertainty quantification, exploration strategies, and theoretically grounded algorithms for decision-making systems. Recent publications highlight expertise in model-based reinforcement learning (e.g., MOMBO for offline RL), PAC-Bayesian bandits, evidential learning for robust classification, and neural stochastic differential equations. His scientific awards include a Best Paper Award (2017) and ELLIS Membership. He also explores interdisciplinary applications of natural sciences to sustainable technology development.

Jocelyn ChanussotView profile

Professor

• Remote Sensing
• Hyperspectral Imaging
• Signal and Image Processing
• +4 more

Jocelyn Chanussot is a Professor at Grenoble Institute of Technology, holding the AXA Chair of Remote Sensing. He is affiliated with GIPSA-Lab (Laboratoire de traitement du Signal et des Images) at Ense3 (École Nationale Supérieure de l'Énergie, l'Eau et l'Environnement) and maintains a connection with the Chinese Academy of Sciences through his AXA Chair position. His research focuses on advancing remote sensing technologies, particularly in hyperspectral imaging and artificial intelligence applications for environmental monitoring. Professor Chanussot specializes in hyperspectral imaging, which captures information across several hundred wavelengths, allowing for detailed characterization of physical properties in observed scenes. His work develops algorithms to extract meaningful information from complex remote sensing data, with applications spanning natural disaster monitoring, environmental observation, biodiversity assessment, and urban planning. He has pioneered approaches that leverage artificial intelligence, particularly deep learning techniques, to process and analyze large-scale remote sensing datasets. His scholarly output shows a clear trend toward integrating advanced AI techniques with remote sensing data. Recent publications demonstrate growing emphasis on transformer architectures, contrastive learning, generative models, and foundation models specifically adapted for hyperspectral data. His work increasingly addresses multimodal data fusion, anomaly detection, and real-time processing for applications in natural disaster response. Ranked among the 157 most cited French researchers in 2019 by Clarivate Analytics Professor Chanussot actively serves on numerous conference program committees, particularly for SPIE's Image and Signal Processing for Remote Sensing conferences. His research is supported by the AXA Research Fund through his AXA Chair in Remote Sensing, which focuses on developing algorithms for natural disaster monitoring and emergency response. He collaborates extensively with international institutions including UCLA and Stanford University on applications ranging from toxic gas detection to tropical forest biodiversity assessment. He leads research at GIPSA-Lab, focusing on developing advanced tools and algorithms for extracting information from complex remote sensing data. His team works on processing heterogeneous data including aerial photos, multispectral and hyperspectral images, and other environmental measurements to improve natural disaster prediction and response capabilities.

• Language Acquisition
• Cognitive Science
• Computational Modeling
• +7 more

Emmanuel Dupoux is a Professor at École des Hautes Études en Sciences Sociales (EHESS), affiliated with the School of Advanced Studies in Social Sciences and the Laboratory of Cognitive Science and Psycholinguistics (LSCP). His work bridges cognitive science, computational linguistics, and machine learning to study infant language acquisition and social cognition. Co-creator and director of EHESS Cognitive Science Master program Former LSCP laboratory director (1998-2009) Current research focuses on textless speech modeling and ecological audio analysis Research areas include: Modeling early language acquisition mechanisms using Bayesian models and HMM Investigating phonological 'deafness' and critical age plasticity Studying social cognition development through infant-toddler experiments His technical contributions span speech processing toolkits like Shennong, emergent communication frameworks (STOP dataset), and innovative approaches to self-supervised speech modeling. Recent publications demonstrate breakthroughs in: Topography-inspired CNN designs for better memory efficiency Prosody-aware generative spoken language models (pGSLM) Textless emotion conversion systems He actively explores how machine learning can reverse-engineer infant language learning processes from ecological audio data, with applications in neurodegenerative disease diagnostics. Key collaborations include: INRIA's Cognitive Machine Learning (CoML) team Facebook AI Research (FAIR) partnerships VoxPopuli multilingual speech corpus development

Grégoire MontavonView profile

Professor

• Explainable AI
• Machine Learning
• Medical Diagnosis
• +2 more

Grégoire Montavon is a Professor at Charité – Universitätsmedizin Berlin and Research Group Lead at the Berlin Institute for the Foundations of Learning and Data (BIFOLD). His appointment commenced on April 1, 2025, as part of BIFOLD's institutional partnership with Charité. He holds a Master's in Communication Systems from École Polytechnique Fédérale de Lausanne (2009) and a Ph.D. in Machine Learning from Technische Universität Berlin (2013). Master's: Communication Systems, EPFL (2009) Ph.D.: Machine Learning, TU Berlin (2013) Montavon pioneers Explainable AI (XAI) for medical applications, developing methods like Layer-Wise Relevance Propagation (LRP) to verify deep learning models in diagnostics. His work bridges machine learning theory with clinical practice, focusing on model transparency for tumor classification, bias mitigation, and regression strategy analysis. Recent research emphasizes counterfactual explainers and spectral analysis for explanation quality assessment. His 2025 publications reveal a cohesive trajectory: advancing explainability frameworks for distance-based classifiers, diffusion models in oncology, and visual counterfactual systems. Key themes include robustness against dataset shifts, metadata integration for fairness, and formal desiderata for explainer design—consistently targeting real-world medical AI deployment. Award highlights: 2013 Dimitris N. Chorafas Award 2020 Pattern Recognition Best Paper Award 2022 Digital Signal Processing Best Paper Award 2025 XAI Conference Best Paper Award (for XpertAI) As BIFOLD Research Group Lead for Explainable Machine Learning in Medicine, Montavon directs projects funded through Charité-BIFOLD partnerships, including the agility project on deep generative model transparency. His team collaborates with Klaus-Robert Müller and others on NIH/DFG grants for AI-driven cancer treatment personalization and Clever-Hans strategy pruning. He leads the BIFOLD research unit Explainable Machine Learning in Medicine, focusing on clinical AI integration. Current initiatives include ICLR 2025 contributions on foundation model reliability and MedI diffusion frameworks for tumor classification bias reduction.