جست‌وجوی استاد و پژوهشگر

Panayotis Mertikopoulosمشاهده پروفایل

استادیار

• Game Theory
• Machine Learning
• Optimization
• +۴ مورد دیگر

Dr. Panayotis Mertikopoulos is a CNRS researcher (chargé de recherche) at the Laboratoire d'Informatique de Grenoble, part of Université Grenoble Alpes. He is affiliated with the Inria/LIG joint team POLARIS and has held visiting positions at UC Berkeley, EPFL, LUISS University of Rome, and NKUA. His academic journey includes completing his PhD at the University of Athens in 2010 on "Stochastic perturbations in game theory and applications to networks" and his Habilitation à Diriger des Recherches (HDR) in 2019 on "Online optimization and learning in games: Theory and Applications". Dr. Mertikopoulos' research spans several interconnected fields at the intersection of mathematics, computer science, and economics. His primary research interests include: Game theory and its applications to network design and resource allocation Online learning algorithms and their convergence properties Optimization methods for non-convex and stochastic problems Applications to machine learning, signal processing, and wireless networks Quantum game theory and quantum computing applications His extensive publication record shows a clear evolution from foundational work in game dynamics and learning theory toward increasingly sophisticated applications in machine learning and network optimization. Recent work demonstrates growing interest in quantum game theory, non-convex optimization, and the mathematical foundations of deep learning. His research consistently bridges theoretical insights with practical applications, particularly in communication networks and distributed systems. Among his notable achievements is receiving the INFORMS best paper award in the network analytics section in 2022 for his work on "Robust power management via learning and game design". His publications have appeared in top venues including NeurIPS, ICML, COLT, IEEE Transactions, and leading economics and operations research journals. Dr. Mertikopoulos has supervised numerous PhD students and postdoctoral researchers, though specific names are not listed in the available information. He has secured research funding for projects at the intersection of game theory, optimization, and machine learning, with applications to network design and resource allocation. His collaborative work spans multiple institutions across Europe and North America. As a member of the POLARIS research team at Inria/LIG, he contributes to a vibrant research environment focused on parallel and distributed systems. His work often intersects with colleagues researching optimization algorithms, machine learning theory, and network science, creating opportunities for cross-disciplinary collaboration on complex computational problems.

Bruce DENBYمشاهده پروفایل

استاد

• Speech Processing
• Silent Speech Interfaces
• Telecommunications
• +۴ مورد دیگر

Bruce DENBY is a Professor at Sorbonne University specializing in speech processing, telecommunications, and indoor localization. His research spans multiple disciplines including computer science, physics, and environmental science with significant contributions across these fields. His primary research interests include: Silent Speech Interfaces and speech restoration technologies Indoor localization using wireless networks and GSM fingerprints Telecommunications and signal processing Environmental modeling of road dust and air pollution Machine learning applications in speech recognition Dr. DENBY's research trajectory shows evolution from early work in high energy physics to speech processing and wireless communications, with recent publications (2022-2025) demonstrating continued innovation in WiFi analytics, client density mapping, and future speech interfaces. His work increasingly integrates deep learning techniques while maintaining focus on practical applications, particularly for speech restoration and privacy-preserving network analysis. Notable scientific achievements: Chester Sall Award Paper (2012) for work on FPGA-based FM broadcast receivers Significant contributions to the Silent Speech Challenge benchmark with deep learning approaches Development of the NORTRIP model for road dust emissions Highly cited work on Silent Speech Interfaces (over 500 citations) Dr. DENBY has secured research funding across multiple domains, collaborating with institutions across Europe. His work demonstrates strong interdisciplinary connections between speech technology, wireless communications, and environmental science, with applications ranging from assistive technologies to urban air quality management.

• Signal and Image Processing
• Bayesian Statistics
• Markov Chain Monte Carlo (MCMC) Methods
• +۴ مورد دیگر

Jean-François Giovannelli is a Professor at IMS Bordeaux , affiliated with the Université de Bordeaux . His work focuses on Signal and Image Processing within the SPECTRAL team. Collaborations span institutions like CEA , CNRS , and industry partners ( STMicroelectronics , Stellantis ), emphasizing Bayesian methods and MCMC algorithms for diverse applications. Current Affiliation : Professor, IMS Bordeaux, Université de Bordeaux Research Group : Signal and Image Processing Team : SPECTRAL Collaborations : CEA, CNRS, CESTA, Thales, NXP Research Interests include: Bayesian inference for inverse problems MCMC sampling techniques Signal/image reconstruction Mass spectrometry data analysis Adaptive optics in astronomy Medical imaging algorithms Article Trends show a focus on Bayesian modeling across disciplines: biomedical data, astronomical imaging, and microwave mapping. Key methods include MCMC samplers , regularized inversion , and statistical validation . Professional Contributions involve developing tools like NiftyRec for tomography and advancing Adaptive Optics restoration algorithms. His work bridges theoretical statistics and practical applications in healthcare, space, and defense.

• Signal Processing
• Image Matching
• Computer Vision
• +۵ مورد دیگر

Guillaume Bourmaud is an Associate Professor at the University of Bordeaux , affiliated with the Bordeaux Institute of Technology and the Signal and Image Processing department. He is a member of the MOTIVE research team within the IMS Bordeaux laboratory. Research Interests : His work focuses on Signal processing and image matching algorithms Computer vision and Transformer-based architectures Statistical learning for geospatial and atmospheric data SLAM (Simultaneous Localization and Mapping) with probabilistic models Publications Trends : Recent work spans computer vision (2024), geospatial navigation (2023), and atmospheric measurement techniques (2022), with emphasis on kriging, Gaussian processes, and feature correspondence evaluation.

• Computer Science
• Signal and Image Processing
• Remote Sensing
• +۵ مورد دیگر

Christian Germain is a Professor of Computer Science at Bordeaux Sciences Agro, an engineering school specializing in agronomy. He focuses on information technologies and their applications to agriculture and environmental science, conducting research in image analysis at the IMS laboratory. His work spans remote sensing, embedded agricultural imaging, and digital tool development for vineyards. Key Roles: Co-holder of the AgroTIC business chair (29 corporate sponsors), Scientific Director of DigiLab (open platform for wine-growing experiments). Research Themes: Remote sensing, agricultural imaging systems, covariance pooling in machine learning, and texture analysis for material science. His recent publications highlight collaborations with industry and academic partners, emphasizing applications in vineyard health monitoring, carbon composite modeling, and vine disease detection. Germain’s team utilizes CNNs, Gaussian mixture models, and SAR imaging techniques to advance agricultural and materials engineering. He has contributed to international conferences and journals, integrating computational methods with real-world agricultural challenges, including proximal sensing for crop management and 3D microstructure simulation.

Cédric HERZETمشاهده پروفایل

پژوهشگر ارشد

• Statistical Learning Theory
• Optimization
• Operational Research
• +۳ مورد دیگر

Cédric HERZET is a Permanent Member of CREST at INRIA Rennes, focusing on statistical learning theory, optimization, and inverse problems. His work bridges theoretical analysis with practical algorithm development. Primary affiliation: INRIA Rennes (French National Institute for Research in Digital Science) Research Interests: Statistical Learning Theory, Optimization algorithms, Operational Research, Inverse Problems. His work particularly addresses sparse approximation, compressed sensing, and iterative thresholding methods. Key Contributions: Development of Bayesian pursuit algorithms, geometric analysis of subspace clustering with outliers, and analysis of state evolution in dense graph message passing. His theoretical work has direct applications in signal/image processing and machine learning. Software: Created MATLAB implementations of sparse approximation algorithms and Bernoulli-Gaussian Lab toolbox for Bayesian pursuit simulation.

• Signal Processing
• Probability
• Inverse Problems
• +۷ مورد دیگر

Christian Heinrich is a Professor at the University of Strasbourg, affiliated with the ICube - IMAGeS team (Images, Modeling, Learning, Geometry and Statistics). His research focuses on signal processing, inverse problems, machine learning, and optical imaging, particularly polarimetric imaging and biological tissue analysis. He collaborates closely with Jihad Zallat (ICube-TRIO) and has contributed to the development of methodologies for medical imaging and computational neuroscience. Education: Electrical Engineering (Supélec, 1990), Master's in Signal Processing (Université Paris-Sud/Supélec, 1994), PhD in Signal Processing (Université Paris-Sud/Supélec, 1997), Aggregation in Applied Physics (1998), Habilitation (University of Strasbourg, 2008) His research spans polarimetric imaging techniques for tumor monitoring, brain connectivity analysis through multimodal MRI and fMRI, and advanced image processing methods like non-local means filtering and spatial mixture models. Recent work explores cryo-EM reconstruction, topological preservation in surgical planning, and uncertainty quantification in medical imaging. Christian Heinrich co-leads the IRIV master’s degree (Imaging, Robotics, and Engineering for the Living) and the Franco-German DaSAI double degree program in Data Science and Artificial Intelligence. He has published extensively in journals like Biomedical Optics Express , IEEE Transactions on Biomedical Engineering , and Medical Image Analysis .

Julie Schollerمشاهده پروفایل

دانشیار

• Statistics
• Data Analysis
• Data Mining
• +۷ مورد دیگر

Julie Scholler serves as an Associate Professor of Mathematics at the University of Tours, working within the Economic Sciences Section of the UFR of Law, Economics and Social Sciences. She holds the position of Head of the first year of the MÉcEn (Master's in Business Economist) program and has been actively teaching since at least 2013 across various levels of the economics curriculum. Her research interests focus on advanced statistical methodologies and data science applications, particularly in exploratory data analysis, factor analysis, unsupervised classification, and machine learning techniques. She specializes in applying these methods using R programming, with expertise in RMarkdown, RShiny, and data visualization techniques. Her work spans from fundamental statistical theory to practical Big Data solutions with R optimization. Analysis of her recent publications reveals a strong emphasis on classification methods, particularly boosting algorithms, random forests, and ensemble methods. Her work demonstrates expertise in prediction error estimation, with applications across various domains including economics, social sciences, and business analytics. She maintains a consistent focus on practical implementation of statistical methods using R, with particular attention to reproducible research practices through RMarkdown. Professor Scholler has developed extensive teaching materials covering statistics, data analysis, and R programming across multiple academic levels. Her courses include Descriptive Statistics, Inferential Statistics, Bayesian Statistics, Dynamical Systems, and specialized topics in Data Mining and Big Data. She appears to be actively involved in student projects and practical applications of statistical methods, as evidenced by student-produced dashboards on topics ranging from world happiness to Covid vaccination data.

El-Ghazali TALBIمشاهده پروفایل

استاد

• Combinatorial Optimization
• Multi-objective Optimization
• Metaheuristics
• +۵ مورد دیگر

El-Ghazali TALBI is a Professor of Exceptional Class at Polytech Lille, University of Lille, where he leads the BONUS research team at Inria Euratechnologies. His career spans over 35 years with continuous academic progression from Researcher to full Professor. He earned a Master in Computer Science from ENSIMAG/INPG (1989), a PhD in Computer Sciences from INPG under T. Muntean (1993) on distributed memory parallel architectures, and a Habilitation from University of Lille 1 (2000) on combinatorial optimization. His academic trajectory includes positions at Grenoble institutions and uninterrupted service at Lille since 1994. TALBI's research centers on combinatorial and multi-objective optimization through metaheuristics, with pioneering work in parallel/distributed algorithms and machine learning integration. His expertise bridges theoretical optimization with industrial applications in bioinformatics, telecommunications, smart grids, and cloud computing, notably advancing Bayesian optimization for neural architecture search and deep learning systems. As a prolific mentor, he has supervised 12 PhD theses including recent defenses on Gaussian processes for aerospace systems (2021), Bayesian neural architecture search (2024), and sustainable power system optimization (2025). His BONUS team develops frameworks like PARADISEO for metaheuristics and GUIMOO for multi-objective evaluation. He has delivered over 100 international invited talks across 20+ countries, with recent focus (2020-2022) on data-driven metaheuristics, neural architecture search, and machine learning optimization synergies. His academic service includes keynote addresses at major conferences including IEEE WCCI, MIC, and ROADEF.

Pierre-Antoine Thouveninمشاهده پروفایل

استادیار

• Signal Processing
• Inverse Problems
• Hyperspectral Image Unmixing
• +۵ مورد دیگر

Pierre-Antoine Thouvenin is an Assistant Professor at Centrale Lille, a prestigious engineering school within the University of Lille community in France. He is a member of the SigMA team from the CRIStAL laboratory (Centre de Recherche en Informatique, Signal et Automatique de Lille), where he conducts research on inverse problems with applications to remote sensing and astronomy. His academic journey began with an Engineering degree in Electronics and Signal Processing from INP - ENSEEIHT Toulouse in 2014, followed by a Master of Science in "Signal, Image, Acoustics" from the same institution. He completed his Ph.D. in "Signal, Image, Acoustics" from Institut National Polytechnique de Toulouse between 2014 and 2017, with research focused on modeling spatial and temporal variabilities in hyperspectral image unmixing. Thouvenin's research interests span several interconnected areas in signal processing and computational imaging. His primary focus is on solving inverse problems, particularly in the context of radio-interferometric imaging for astronomy and hyperspectral image unmixing for remote sensing applications. He has made significant contributions to developing advanced algorithms for handling spectral variability in hyperspectral data and for image reconstruction in radio astronomy. His work often combines Bayesian statistical methods with optimization techniques to address challenging high-dimensional problems. A distinctive aspect of his research is the development of distributed and parallel computational methods that enable processing of extremely large datasets that would be intractable with conventional approaches. An analysis of his recent publications reveals a strong trend toward developing distributed and parallel computational methods for large-scale inverse problems. His work increasingly integrates machine learning approaches, particularly neural networks, with traditional signal processing techniques. There's a clear progression from theoretical developments in hyperspectral unmixing to practical applications in astronomy, particularly through his involvement in the ORION-B project where he applies statistical methods to infer physical conditions in star-forming regions. His most recent work demonstrates sophisticated integration of spatial regularization techniques with Bayesian inference for astrophysical parameter estimation. Prix Léopold Escande from Institut National Polytechnique de Toulouse (2017) - awarded to the best PhD theses defended at INPT Prix de l'Institut National Polytechnique de Toulouse (2014) - awarded for outstanding academic achievement during engineering studies Thouvenin is actively involved in mentoring the next generation of researchers. He currently co-supervises the PhD thesis of Pierre Palud on "Statistical methods for model inversion and spatial distribution of physico-chemical properties of the molecular cloud Orion B" as part of the CNRS 80|Prime project OrionStat. His research is supported through various academic collaborations and projects, including the ORION-B project led by Jérôme Pety, which involves molecular line observations from the IRAM-30m Large Program. He has established productive international collaborations, particularly with researchers at Heriot-Watt University in Edinburgh where he worked as a Research Associate from 2017-2019. Thouvenin is a key member of the SigMA team within the CRIStAL laboratory, a joint research unit between Centrale Lille, INRIA, and University of Lille. His work often intersects with the ORION-B project, where he collaborates with astrophysicists to develop statistical methods for inferring properties of Galactic and extra-galactic star forming regions. This interdisciplinary environment fosters innovation at the intersection of signal processing, statistics, and astronomy. He has developed expertise in translating complex statistical methodologies into practical computational tools that address real-world challenges in both remote sensing and astronomical imaging.