Find Academic Supervisors

• Reinforcement Learning
• Computational Decision-Making
• Artificial Intelligence
• +5 more

Carlo D'Eramo is Professor and Head of the Reinforcement Learning and Computational Decision-Making professorship at the Center for Artificial Intelligence and Data Science (CAIDAS) of University of Würzburg. He additionally serves as an independent group leader of hessian.AI, focusing on developing lightweight methods to obtain adaptive autonomous agents that can handle real-world complexity. His academic journey includes: B.Sc. in Computer Engineering from Politecnico di Milano (2011) M.Sc. in Computer Engineering from Politecnico di Milano (2015) Double degree in Computer Science from University of Illinois at Chicago (2015) Ph.D. in Information Technology from Politecnico di Milano (2019) Postdoctoral research at TU Darmstadt's Intelligent Autonomous Systems group (2019-2022) D'Eramo leads the LiteRL research group investigating how agents can efficiently acquire expert skills accounting for real-world complexity. His research spans multiple reinforcement learning domains including multi-task, curriculum, adversarial, options, and multi-agent RL. His work bridges theoretical advances with practical applications, particularly in robotics and decision-making systems. His recent publications (2023-2025) demonstrate significant contributions across exploration strategies, neural network architectures, multi-agent coordination, and physics-informed machine learning. His work frequently appears in top-tier venues including TMLR, RLJ, IEEE PAMI, ICML, and ICLR, with multiple papers receiving spotlight or oral presentation designations. Professional activities include: Senior area chair for RLC Area chair for AAAI, ACML, AISTATS, NeurIPS, and ICLR Reviewer for DFG and ERC proposals Creator of MushroomRL reinforcement learning framework D'Eramo actively contributes to academic community service while mentoring researchers in his group. His work on lightweight methods aims to make reinforcement learning more practical for real-world applications across various domains.

Tim Ruben DavidsonView profile

Researcher

• deep generative models
• agentic systems
• synthetic data
• +1 more

Tim Ruben Davidson is a Researcher at the Digital Life Lab (DLAB) within the School of Computer and Communication Sciences at EPFL. His current role is Doctoral Assistant, pursuing a doctoral program in Computer and Communication Sciences. His research focuses on advanced machine learning topics including deep generative models, agentic systems, synthetic data applications, and representation learning. His work bridges theoretical advancements in AI with practical implications, particularly in understanding AI agency, synthetic data generation, and latent space structures. Key research trends in his articles include exploring AI self-awareness, evaluating AI-driven peer review systems, and optimizing generative models through geometric and topological approaches. He has contributed to foundational studies on hyperspherical VAEs and reparameterization techniques on Lie groups, advancing mathematical foundations of neural networks. His work is published in top-tier venues and reflects interdisciplinary engagement between computer science, mathematics, and ethical AI considerations.

Francesca Dalia FaraciView profile

Senior Lecturer

• Biomedical Signal Processing
• Medical Data Analysis
• Wearable Technology
• +5 more

Francesca Dalia Faraci is a Senior Lecturer and Researcher at the Department of Innovative Technologies (DTI) of SUPSI. She holds a PhD in Electronics from the University of York (UK) and a Laurea in Physics from the University of Genoa (Italy). Her research focuses on biomedical signal processing, statistical data analysis, and wearable technology applications in healthcare. She leads projects such as Cardio MIPA (arrhythmia monitoring) and AutoPlay_DD (child development analysis via smart toys). Key contributions include developing algorithms for sleep staging, arrhythmia detection, and synthetic ECG generation. Awards include the GOSPEL and DoE School Certificates. She coordinates research projects and teaches courses in Physics and Applied Statistics. Education: PhD in Electronics (University of York, 2006), Laurea in Physics (University of Genoa, 2002) Roles: Senior Lecturer, Researcher, Project Coordinator Key Projects: Tackling teleWorking Impacts, my Doctor's Lifestyle (lifestyle medicine), Telemonitoring Breast Cancer Her research bridges AI and clinical practice, emphasizing ethical frameworks and complementarity between clinicians and AI systems. Recent work includes causal inference for sleep dynamics and bias quantification in healthcare algorithms.

Julian TachellaView profile

Researcher

• Signal Processing
• Machine Learning
• Computational Imaging
• +5 more

Dr. Julian Tachella is a CNRS Research Scientist at the Sisyph Laboratory of École Normale Supérieure de Lyon, with co-founder/CSO roles at Blur Labs. His career spans signal processing, machine learning, and computational imaging, focusing on inverse problems and self-supervised learning. Affiliation: CNRS (French National Centre for Scientific Research), Sisyph Laboratory, École Normale Supérieure de Lyon Co-founder & CSO: Blur Labs (AI/Imaging startup) Research Interests: At the intersection of signal processing and deep learning , his work addresses imaging inverse problems through self-supervised methodologies (e.g., UNSURE, Generalized R2R) that eliminate ground-truth requirements. Key contributions include equivariant imaging frameworks for stability, spline sketches for photon-counting lidar compression, and uncertainty quantification techniques with equivariant bootstrapping. Recent Trends: 2025 publications emphasize lightweight architectures for multi-domain reconstruction (CT, super-resolution) and noise-agnostic SURE methods. 2024 works focus on audio declipping , compressed lidar , and nonlinear algorithm unrolling with applications in autonomous vehicles and medical imaging. Scientific Awards: Best Student Paper Award at ICASSP’22 Collaborations & Leadership: He leads the DeepInverse open-source project and develops algorithms for real-time 3D lidar reconstruction. His team includes researchers from University of Edinburgh and Grenoble INP, with applications in automotive lidar and underwater imaging.

Roberto CalandraView profile

Professor

• Robotics
• Machine Learning
• Tactile Sensing
• +3 more

Roberto Calandra is a Full (W3) Professor at Technische Universität Dresden, where he leads the Learning, Adaptive Systems and Robotics (LASR) Lab. Previously, he served as a Research Scientist at Meta AI (formerly Facebook AI Research) and a Postdoctoral Scholar at UC Berkeley’s BAIR Lab under Sergey Levine. His academic journey includes a PhD in Robotics from TU Darmstadt, an M.Sc. in Machine Learning from Aalto University, and a B.Sc. in Computer Science from Università di Palermo. His research bridges Robotics and Machine Learning, focusing on tactile sensing, Bayesian Optimization, and model-based reinforcement learning. He pioneered the DIGIT tactile sensor , now the most widely used tactile sensor in robotics, and advocates for a computational field of Touch Processing to advance haptic understanding. His work emphasizes data-efficient learning, real-world dexterous manipulation, and multimodal perception. Recent publications highlight breakthroughs in tactile sensor design, in-hand object manipulation, and multimodal integration. He organizes workshops on robotics and machine learning (e.g., at NeurIPS, ICRA) and promotes open-source tools like PyTouch and TACTO .

Andreas KrauseView profile

Professor

• Machine Learning
• Artificial Intelligence
• Reinforcement Learning
• +5 more

Andreas Krause is a Professor of Computer Science at ETH Zurich, leading the Learning & Adaptive Systems Group and serving as Chair of the ETH AI Center and Academic Co-Director of the Swiss Data Science Center. He previously held an Assistant Professorship at Caltech. His research focuses on machine learning, Bayesian optimization, reinforcement learning, and AI safety. Key contributions include advancements in active sensing, learning-based control, and probabilistic AI. Affiliations: ETH AI Center, Swiss Data Science Center, Max Planck Institute for Intelligent Systems (Fellow) Education: PhD in Computer Science (Carnegie Mellon University, 2008); Diplom in Computer Science and Mathematics (Technical University of Munich, 2004) His work bridges theory and practice, with applications in robotics, healthcare, and climate science. Notable awards include the IEEE Fellow distinction, Rössler Prize, and multiple ERC grants. Krause actively chairs ICML conferences and contributes to global AI policy through the UN’s High-level Advisory Body on AI. Research Interests: Machine learning fundamentals, adaptive systems, AI safety, and optimization. Recent work explores safe exploration in reinforcement learning, causal modeling, and large-scale generative AI.

• Geotechnics
• Numerical Simulations
• Bayesian Inference
• +4 more

Stéphane Commend is an Associate HES Professor at the Fribourg School of Engineering and Architecture (HEIA-FR) under HES-SO Valais-Wallis. He also holds a lecturer role at the School of Engineering and Management of the Canton of Vaud. His primary research focuses on geotechnics, numerical simulations, and probabilistic modeling applied to infrastructure projects like tunneling and deep excavations. Education and affiliations include roles across multiple HES-SO institutions, with a strong emphasis on integrating advanced computational methods into geotechnical engineering. Notable projects include the Grand Paris Express tunnel project, Bayesian inference for wood constitutive modeling, and probabilistic risk analysis for urban construction. Research interests span soil-structure interaction, finite element modeling, and uncertainty quantification. Recent work emphasizes Bayesian methods for parameter calibration, machine learning in excavation design, and natural hazard vulnerability assessment. Key contributions include frameworks linking ZSOIL and UQLab for reliability analysis, and prototypes like SLIDE-PM for mudflow impact modeling. Current projects (e.g., iBAG and OptiSoil) focus on optimizing construction methods using AI and data-driven approaches. He leads collaborative teams across HES-SO institutes and academic partners like EPFL and CETU. Key Projects: iBAG Project (2022–2025): Bayesian methods in geotechnics OptiSoil (2019–2025): Machine learning for excavation design TULIP Project: TBM-pile interaction probabilistic analysis

• Applied Analysis
• Numerical Analysis
• Data Science
• +6 more

Leon Bungert is a Professor of Mathematics of Machine Learning at the University of Würzburg, working in applied analysis and numerics with a particular focus on data science and machine learning. His research investigates PDEs and variational models on graphs, adversarial robustness of machine learning, variational regularization, and nonlinear optimization. Dr. Bungert serves as a guest editor for the European Journal of Applied Mathematics, an associate editor for Advances in Continuous and Discrete Models: Theory and Applications, and is a member of the program committee at SSVM 2025. He is also an ELLIS member and actively organizes conferences and workshops, including "MIA'25" at IHP in Paris (January 13-15, 2025), "Synergies of Machine Learning and Numerics" in Osaka (March 11-13, 2025), and "Mathematical Analysis of Adversarial Machine Learning" in Oaxaca (August 17-22, 2025). Research Interests Dr. Bungert's primary research areas include: PDEs on graphs Adversarial robustness in machine learning Inverse problems Optimization Variational problems in L-infinity Nonlinear eigenvalue problems Image reconstruction with structural priors His work bridges theoretical mathematics with practical applications in machine learning, particularly focusing on the mathematical foundations of deep learning and developing robust algorithms that can withstand adversarial attacks. He has made significant contributions to understanding the connections between partial differential equations and machine learning algorithms. Research Trends Analysis of Dr. Bungert's recent publications reveals a strong focus on the intersection of machine learning and mathematical analysis. A key theme is the application of variational methods and partial differential equations to machine learning problems, particularly in understanding and improving the robustness of neural networks against adversarial examples. His work on Lipschitz learning on graphs has established important theoretical foundations for graph-based semi-supervised learning. Additionally, his research on the infinity Laplacian and p-Laplacian equations provides deep insights into the mathematical structure of machine learning algorithms. The development of Bregman learning frameworks for sparse neural networks represents a significant contribution to efficient deep learning model training. Professional Activities Dr. Bungert is actively involved in the academic community through editorial roles and conference organization. His current professional activities include: Guest editor for the European Journal of Applied Mathematics Associate editor for Advances in Continuous and Discrete Models: Theory and Applications Member of the program committee at SSVM 2025 ELLIS member Co-organizer of multiple international conferences and workshops Technical Contributions Dr. Bungert has developed several open-source software packages that implement his theoretical contributions, including: Code for convergence rates of Lipschitz learning on graphs A Bregman training framework for sparse neural networks CLIP: Cheap Lipschitz Training of Neural Networks Nonlinear Power Method for Proximal Operators and Neural Networks Robust Image Reconstruction with Misaligned Structural Information These implementations are primarily in Python and MATLAB, demonstrating his commitment to making theoretical advances accessible for practical applications.

Umberto MichelucciView profile

Professor

• Machine Learning
• Artificial Intelligence
• Astrophysics
• +5 more

Umberto Michelucci is a Professor of Scientific Machine Learning at Lucerne University of Applied Sciences and Arts (HSLU), Switzerland. He holds a PhD in Machine Learning applied to Physics and has over 20 years of industry experience. He is the Subject Head of Applied Data Intelligence in Continuing and Executive Education, Head of Certificates in Machine Learning/Data Engineering, and founder of TOELT LLC and the AI Center of Excellence at Helsana Versicherung AG. His research focuses on machine learning applications in science, astrophysics, uncertainty quantification, and sensor technology. Education PhD in Machine Learning applied to Physics (Portsmouth University) Master in Theoretical Physics (University of Florence) Postgraduate Certificate in Higher Education (Open University, UK) Research Interests Michelucci’s work bridges machine learning and scientific disciplines. Key areas include: Machine learning for astrophysics (INAF collaborations) Uncertainty analysis in high-stakes ML systems Deep learning for optical sensing (e.g., olive oil quality analysis) Foundational mathematical concepts for ML in science Awards & Recognition World’s Top 2% Scientists (Stanford List) Google Developer Expert in Machine Learning AI Global Ambassador (2022) TOP AI Influencer in Switzerland (2021) Grants & Collaborations He collaborates with institutions like INAF (Italy) and NVIDIA/Google, leading projects on AI for agrifood, medical imaging, and astrophysics. His work includes $multi-million industry partnerships and EU-funded research. Labs & Teams Director of the TOELT AI Lab and oversees HSLU’s Applied Data Intelligence programs. Active in open-source initiatives and global AI standardization efforts.

• Geometric Deep Learning
• Topological Deep Learning
• Topological Data Analysis
• +7 more

Bastian Alexander Grossenbacher is a Full Professor of Machine Learning at the University of Fribourg, where he leads the AIDOS Lab within the Department of Informatics, Faculty of Science and Medicine. He holds secondary appointments at the Institute of AI for Health and the Helmholtz Pioneer Campus of Helmholtz Munich. He is also a TUM Junior Fellow and a member of ELLIS and AI-LIFE. He co-directs the Applied Algebraic Topology Research Network (AATRN) and is involved in the Topology, Algebra, and Geometry in Data Science (TAG DS) initiative. His research lies at the intersection of geometry, topology, and machine learning, with a focus on Geometric Deep Learning, Topological Deep Learning, and Topological Data Analysis. He develops novel machine learning methods that use topological and geometric priors to improve model robustness and interpretability, particularly in biomedical applications. His work spans theoretical foundations and practical implementations in areas such as single-cell analysis, medical imaging, and network science. The 15 most recent publications (2023–2025) reveal a strong trend toward integrating algebraic and differential topology into deep learning architectures. Key themes include the use of Euler Characteristic and magnitude transforms, curvature-based analysis of graphs and data, simplicial and manifold-based neural representations, and the application of topological methods to biomedical data. His articles appear in top-tier venues such as Nature Communications, ICML, NeurIPS, ICLR, and IEEE conferences, reflecting both theoretical depth and high-impact applications. ERC Starting Grant Bastian Rieck is actively involved in mentoring and academic leadership. He supervises PhD students and postdoctoral researchers in the AIDOS Lab and is open to new collaborations, particularly with candidates from interdisciplinary backgrounds. He has secured significant research funding, including the ERC Starting Grant, and leads multiple collaborative research initiatives. He emphasizes open science, sharing code, data, and educational materials publicly. He leads the AIDOS Lab, which focuses on advancing machine learning through topological and geometric principles. He is also a co-director of the Applied Algebraic Topology Research Network (AATRN), which hosts a large online seminar series. Additionally, he maintains DONUT, a curated database of non-theoretical applications of topology, and is active in the ELLIS and AI-LIFE networks, fostering international collaboration in AI and health.

• Medical Data Science
• Anomaly Detection
• Generative AI
• +3 more

Dr. Thomas Sutter is a Researcher affiliated with the Department of Computer Science at ETH Zurich , specifically part of the Professorship for Medical Data Science. His work focuses on advanced machine learning techniques applied to medical data, including anomaly detection, generative AI, and multimodal learning. He contributes to healthcare innovation through projects like improving radiology diagnostics via Vision-Language Models (RadVLM) and developing denoising techniques for physiological signals in cardiology. Research Interests: Medical Data Science, Anomaly Detection in Healthcare, Generative AI for Biomedical Signals, Multimodal Representation Learning, and Cardiac Function Prediction using Echocardiograms. His methods often combine deep learning with domain-specific medical challenges. Key Publications Trends: Recent work emphasizes medical imaging analysis (e.g., MIMIC-CXR studies), generative models for signal denoising, and contrastive learning for anomaly detection. His research bridges computer science theory with clinical applications, particularly in cardiology and radiology. Awards & Grants: No specific awards or grants mentioned in the provided texts. Advising & Teams: While no advisees are listed, he collaborates within the Medical Data Science research group at ETH Zurich, contributing to interdisciplinary projects in healthcare AI.

• Speech Processing
• Automatic Speech Recognition
• Auditory Modeling
• +3 more

Philip Neil Garner is a Researcher at the Idiap Research Institute (LIDIAP), affiliated with École Polytechnique Fédérale de Lausanne (EPFL). His current position is listed as "EPFL member Current" with email philip.garner@epfl.ch. He maintains an active research profile with publications spanning from 2013 to 2025. Garner's research focuses on the intersection of physiological auditory modeling and machine learning for speech processing. His work bridges cochlear physiology with automatic speech recognition systems, investigating how biological principles of hearing can inform and improve computational models. Key areas include modeling the cochlea as an active amplifier using Hopf oscillators, exploring neural oscillations in speech perception via spiking neural networks, and developing interpretable affective speech synthesis systems. His research demonstrates consistent interest in creating biologically plausible models that maintain compatibility with modern deep learning frameworks. Analysis of his recent publications reveals a clear trajectory toward integrating physiological auditory models with state-of-the-art speech recognition systems. His work increasingly focuses on creating hybrid models that maintain physiological plausibility while leveraging pre-trained acoustic models. The research shows particular attention to modular approaches that allow different components (cochlear models, neural networks) to interact meaningfully, with emphasis on understanding how end-to-end learning affects physiological interpretations of speech processing. Garner has supervised multiple doctoral students including Louise Coppieters De Gibson, Bastian Schnell, and Sibo Tong, whose theses address cochlear modeling, affective speech synthesis, and multilingual speech recognition respectively. His research has received funding from the Swiss National Science Foundation as indicated in two publications. While specific grants aren't detailed, his work demonstrates consistent collaboration with Hervé Bourlard and Alexandre Bittar across multiple projects. As a core researcher at Idiap Research Institute, Garner works within a multidisciplinary team focused on speech processing and artificial intelligence. His publications indicate collaboration across units including LIDIAP, EDEE, STI, IEL, and LCAV at EPFL, suggesting integration within both the Idiap institute and broader EPFL research ecosystem. His work connects computational neuroscience with practical speech technology applications, positioning him at the intersection of theoretical auditory modeling and applied speech processing.

• Multiscale/multiphysics modeling
• Biophysics
• Cellular and molecular biomechanics
• +6 more

Igor V. Pivkin is a full professor at the Institute of Computing within the Faculty of Informatics at the University of Lugano (USI). He holds a B.Sc. and M.Sc. in Mathematics from Novosibirsk State University, followed by an M.Sc. in Computer Science and a Ph.D. in Applied Mathematics from Brown University. Before joining USI, he was a Postdoctoral Associate at MIT's Department of Materials Science and Engineering. His research focuses on multiscale/multiphysics modeling , numerical methods, and large-scale simulations of biological and physical systems. Key areas include biophysics, cellular/molecular biomechanics, stochastic modeling, and coarse-grained molecular simulations. He leverages high-performance computing and particle-based methods to address complex biological phenomena. His work spans diverse applications, including cancer cell dynamics, bioleaching bacterial biofilms, and erythrocyte mechanics in human spleen circulation. He has pioneered computational tools such as the Bayesian recursive global optimizer (BaRGO) and the in-silico lab-on-a-chip framework, enabling petascale simulations of microfluidic systems at cellular resolution. Pivkin collaborates extensively with institutions like the SIB Swiss Institute of Bioinformatics and has contributed to advancing methodologies for multi-model scientific simulations. His research integrates experimental data with computational models to bridge gaps between microscopic and macroscopic biological processes.

Ye HongView profile

Researcher

• Geoinformatics
• Urban Planning
• Transportation Engineering
• +3 more

Dr. Ye Hong is a Researcher affiliated with the Institute of Cartography and Geoinformatics at ETH Zürich, specifically within the Department of Geoinformation Engineering. Their work focuses on geospatial data analysis, urban mobility modeling, machine learning applications in transportation systems, and sustainable urban planning. They contribute to open-source tools like Trackintel for mobility analysis and have published extensively on topics such as mobility data synthesis, traffic prediction, and privacy-preserving techniques. Research interests emphasize integrating multi-source geospatial data with deep learning to address challenges in urban sustainability, poverty reduction, and transportation efficiency. Their articles highlight innovations in trajectory generation, uncertainty quantification, and contextual-aware neural networks for spatial-temporal prediction. Dr. Hong’s publications explore the interplay between urban infrastructure, human behavior, and environmental impact. Key themes include accessibility measurement in cities, carbon footprint analysis, and the ethical implications of tracking mobility data. Their work bridges theoretical advancements in geoinformatics with practical applications in urban policy-making and infrastructure design. Notable contributions include frameworks for causal intervention in mobility data, methods to estimate poverty reduction efficiency using remote sensing, and analyses of tracking duration effects on location privacy. These efforts demonstrate a commitment to leveraging geospatial technologies for socially impactful research.

Mutian HeView profile

Academic

• Spoken Language Understanding
• Speech Synthesis
• Natural Language Processing
• +12 more

Mutian He is a PhD candidate and Doctoral Assistant at the École Polytechnique Fédérale de Lausanne (EPFL), Switzerland, affiliated with the Idiap Research Institute and the School of Engineering. He is pursuing his doctoral studies in Electrical Engineering under the supervision of Phil Garner. He holds a B.E. from Beihang University (BUAA) and an MPhil from the Hong Kong University of Science and Technology (HKUST). B.E., Beihang University (BUAA), 2019 MPhil, Hong Kong University of Science and Technology, 2022 PhD Candidate, École Polytechnique Fédérale de Lausanne (EPFL), ongoing His research focuses on spoken language understanding, speech synthesis, and the intersection of speech and language processing with machine learning. He explores efficient model architectures, pretraining strategies, multilingual and low-resource modeling, and the use of large language models in speech tasks. His work spans both theoretical and applied aspects, including distillation to linear-complexity models, robust TTS, and commonsense reasoning via conceptualization. His recent publications at top venues such as ICLR, EMNLP, Interspeech, and KDD demonstrate a strong trend towards efficient and scalable models for speech and language, with increasing emphasis on multilingualism, knowledge transfer, and real-world deployment in low-resource settings. He has also contributed to open-source implementations and community tools like Speech Rankings. Joint Fine-tuning and Conversion of Pretrained Speech and Language Models towards Linear Complexity (ICLR 2025) Acquiring and Modelling Abstract Commonsense Knowledge via Conceptualization (AIJ 2024) The Interpreter Understands Your Meaning: End-to-end Spoken Language Understanding Aided by Speech Translation (Findings of EMNLP 2023) Can ChatGPT Detect Intent? Evaluating Large Language Models for Spoken Language Understanding (Interspeech 2023) Multilingual Byte2Speech Models for Scalable Low-resource Speech Synthesis (2022) Mutian He has served as a teaching assistant for courses including Introduction to Natural Language Processing at HKUST and Introduction to Speech Processing at Idiap. He has also worked on speech synthesis at Microsoft, focusing on robustness and multilingual conditions. He is actively involved in research advising under Phil Garner and has collaborated with multiple researchers across institutions. He is affiliated with the LIDIAP (Laboratory of Intelligent Data Analysis and Pattern Recognition) at EPFL, where he contributes to research in deep learning for speech and language. His lab work involves developing novel neural architectures, conducting experiments on multilingual datasets, and open-sourcing code to promote reproducibility.