Zepeng ZhangZepeng is a PhD student at École Polytechnique Fédérale de Lausanne (EPFL). He works in the Intelligent Maintenance and Operations Systems (IMOS) Lab under the supervision of Prof. Olga Fink. Previously, He received his master’s degree at ShanghaiTech University and B.Eng. degree from Wuhan University (WHU). During his studies, he have visited Télécom Paris, Institut Polytechnique de Paris (IP Paris), Peking University (PKU), and City University of HongKong (CityU) as a research intern. His current research
Chenghao XuEducation Bachelor of Engineering | Mechnical Engineering 2016 – 2020 Southern University of Science and Technology, China Master of Science | Robotics 2020 – 2023 Delft University of Technology, Netherlands
Keivan Faghih NiresiExpertise
Computational sensing/imaging, Inverse problems, Signal processing, Graph neural networks, Domain adaptation.
Keivan is a PhD student at école Polytechnique Fédérale de Lausanne (EPFL). He joined the Intelligent Maintenance and Operations Systems (IMOS) Lab under the supervision of Prof. Olga Fink in February 2023.Prior to joining EPFL, he obtained his master's degree from the Institute of Communications Engineering, College of Electrical Engineering and Computer Science, National Tsing Hua Univer
Vinay SharmaExpertise
Computational Mechanics, Graph Machine Learning.
Sergei GarmaevResearchPhysics-informed deep learning Operator learning, PINNs, dynamical systems modelingPredictive maintenance Estimation of health indicators, remaining useful life
Han SunExpertise
Deep Learning, Computer Vision, Domain Adaptation, Generative Modeling
Open Projects:
Test-Time Domain Adaptation for Anomaly Detection: SiROP link for detailed information
Han obtained her master's degree in Geomatic from the Federal Institute of Technology in Zürich (ETHZ). Before joining IMOS Lab - EPFL, she focused on computer vision with a particular interest in domain adaptation & style adaptation. During her Ph.D. study, she will explore more in the field of domain adaptation with its appli
Amaury Pierre Jiezhi WeiExpertise
Deep Learning, Computer Vision, Physics Informed Machine Learning, Graph Neural Networks
Amaury obtained his Master's degree in Robotics from the Federal Institute of Technology in Lausanne (EPFL) in 2021.Before joining IMOS, he worked for multiple years at Logitech Europe S.A. as a Software Engineer in the field of Augmented and Virtual Reality. He researched 6DoF tracking solutions for handheld devices, developed the firmware and designed the testing strategy for the MR Stylus MX Ink.During his
Olga FinkTeaching & PhD PhD Students Han Sun, Leandro Von Krannichfeldt, Keivan Faghih Niresi, Zepeng Zhang, Vinay Sharma, Chenghao Xu, Raffael Pascal Theiler, Kevin Antonio Steiner, Sergei Garmaev, Amaury Wei Past EPFL PhD Students Ismail Nejjar, Mengjie Zhao Courses Data Science for infrastructure condition monitoring CIVIL-332 The course will cover the relevant steps of data-driven infrastructure condition monitoring, starting from data acquisition, going through the steps pre-processing of real
Mengjie ZhaoMengjie Zhao holds degrees in Computational Mechanics (MSc) with honors track and in Engineering Science (BSc) from the Technical University of Munich (TUM).
From the early years of her studies, Mengjie was fascinated by the modeling of multiphysics and multiscale systems. As a student research assistant at TUM and research intern at International Centre for Numerical Methods in Engineering (CIMNE), she gained a solid understanding of both the theoretical and algorithmic fundamentals as well as a wide range of applications. Through the BGCE project with the Elitenetzwerk Bayern (ENB), which dealt with the mesh sensitivity prediction with a deep neural network, she realized that leveraging data could bring physical modeling far beyond the current computational limits. Later, in her master's thesis in cooperation with Siemens, she turned to the reduced-order modeling with enforced physical invariants, which showed better accuracy and generality.
In her Ph.D., she would like to step towards a further combination of deductive research (modeling and simulation) and inductive (data-driven) research by embedding physics into machine learning.
