Statistical Learning Research Group (Sailing)
Professor in Data Science. PhD degree in Computer and Communication Sciences from EPFL. Stephan Robert was born in Neuchâtel, Switzerland. He graduated as an engineer (BSc) in microengineering with first honors (valedictorian). He received the ing. dipl. degree (MSc) in electrical and electronics engineering (Maillefer Award) and the « Docteur ès Sciences » degree (PhD) with Prof. Jean-Yves Le Boudec, both from EPFL, Lausanne (Switzerland). From 1996 to 1998, he has been a Post-doctoral Research Fellow at UC Berkeley with Prof. Jean Walrand. He worked as a researcher, consultant, engineer and projects manager for many companies during 8 years: Swisscom (Bern), Huber & Suhner (Herisau), ABB (Turgi), GDA (Bern) and Cisco Systems (San Jose, California). Since 2001, he has been a Professor at the Haute Ecole d’Ingénierie et de Gestion du Canton de Vaud (HEIG-Vd / HES-SO), Yverdon. On the political side he has been a « député » (deputy) at the « Grand Conseil Neuchâtelois » (2005-2009). During his Master’s thesis he developed the first prototype of an SSFIP antenna which is manufactured by Huber+Suhner and widely spread over our country. During his PhD, he discovered new models of Markov chains exhibiting self-similarities over a finite timescales. At Swisscom, he was the project leader of « Mobile Unlimited » (which won the highest application distinction, the GSM award in Cannes, March 2005) during the initiation phase. He was also involved in a number of EU projects for traffic engineering, mobility and wireless communications (UWB) aspects. Now his research interests are on artificial intelligence for acute medicine, machine learning, prediction, time series, scheduling, fraud detection, medical emergencies. He also served as an editor for IEEE Communications Surveys and Tutorials and as a TPC member of many conferences. During eleven consecutive years he organized a summer university with American universities and with the competitive Seoul National University (SNU). LinkedIn. Avisdexperts.
Research Associate (Adjoint Scientifique), PhD in Operational Research, Imperial College, UK. MSc in finance University of London. BA and MA in mathematics, University of Cambridge. Dr. Efstratios Rappos (Stratos) is a postdoc at HEIG-VD. He has obtained a Ph.D. in combinatorial optimization / operations research from Imperial College London, a BA and MA in Mathematics from the University of Cambridge and an MSc in Finance from the University of London. He got the Bronze medal at the International Mathematical Olympiads (IMO, 1995). His research interests include mathematical modeling, integer programming, network design, combinatorial optimization, algorithm design and large-scale dataset manipulation. He has worked as a scientific team leader for the UK Department for Work and Pensions (DWP) developing mathematical models for large-scale data mining, cleaning and analysis of social security data, and has also collaborated in several academic EU projects. Recently he won the second place in the International Timetabling competition (DIPLOMA).
Félicien holds a MSc in mathematics with a minor in computer science from the University of Neuchâtel. Passionate about applied mathematics, machine learning, deep learning, and data science, he wrote his master thesis on persistent homology for topological data analysis, an approach which has been developed to analyze complex datasets. Persistent homology is a method for computing topological features of a space at different spatial resolutions. More persistent features are detected over a wide range of spatial scales and are deemed more likely to represent true features of the underlying space rather than artifacts of sampling, noise, or particular choice of parameters. To find the persistent homology of a space, the space must first be represented as a simplicial complex. A distance function on the underlying space corresponds to a filtration of the simplicial complex, that is a nested sequence of increasing subsets. The mathematical tools used in persistent homology mostly come from commutative algebra, combinatorics, and probability theory. Topological data analysis has a wide range of applications, among which image compression, cancer research, and shape or pattern recognition. Currently, he is involved in an exciting European research project in collaboration with the CHUV (Centre Hospitalier Universitaire Vaudois), the University of Franche Comté, the CHU (Centre Hospitalier Universitaire) Besançon, TechWan (www.techwan.com) and Maincare (www.maincare.com), and he tries to find optimal solutions using deep learning to guide and involve the most appropriate vector (helicopter, ambulance) in the face of an emergency, accident, or disaster.
Dr. Denis LEVCHENKO
Denis acquired his BSc in theoretical physics (Franz Mandl Prize) from the University of Manchester and then did his master’s in mathematics (aka Part III Maths) at the University of Cambridge. He continued with mathematics for his PhD at the University of Zurich, where he researched algebraic geometry, specifically problems involving Deligne-Mumford stacks — abstract mathematical objects well-suited to reflecting local symmetries. In between his master’s and PhD, Denis interned at GSK as a data scientist, applying topological data analysis and machine learning (random forests) algorithms to biological 2D and 3D images. He is currently a Research Scientist at HEIG-VD, working in close collaboration with Predictive Layer on a project set to explore neural architecture search (NAS) — an automated process to find the most suitable architecture for a deep neural network — in the context of time series data.
Clément is an experienced professional with 10 years of teaching experience in mathematics, statistics, machine learning, and programming. He holds a Teaching Diploma for secondary 2 (DEEM) and a Master in Mathematics from the University of Fribourg (Switzerland). Clément’s technical and creative skills and his ability to collaborate have enabled him to become an expert in health insurance mathematics and a qualfied cryptologist of the Swiss army. He has a strong sense of consistency and attention to detail, which is evident in his work and commitment to his students’ success.
Dr. Christopher HEMMENS
Chris studied mathematics at Imperial College, London and earned a PhD in finance from the University of Geneva. For the last five years he’s been learning Python and machine learning whilst working as a data science consultant. In addition to his professional work, Chris is a committed thespian who also writes and directs theatre in Lausanne.
Dr. Shabnam Ataee
Shabnam Ataee obtained her master and bachelor degrees in Computer Engineering in Tehran, Iran. She received her Ph.D. in Information Systems from University of Lausanne, Switzerland in 2015. Shabnam was working as a computer scientist and a data scientist in the industry since then. Starting from February 2020, she is working as “Collaboratrice Ra&D HES” at IICT, HEIG-VD. During this period, she had this opportunity to work as “senior data scientist” in different biological, industrial and academic projects to solve different problems, applying different feature selection/ feature engineering techniques, using various machine learning/deep learning models, transfer learning and fine tuning. During this period, she also served as teaching assistant and lab instructor for various data science courses. Her research interests are applied data science, data analysis and solving problems by applying machine learning and deep learning models in healthcare projects.
Anna graduated of her BSc in mathematics at Universitat Autònoma de Barcelona. After her theoretical formation and also having interest in computer science, she decided to focus her career on Artificial Intelligence. She started her MSc in Artificial intelligence at Universitat Politècnica de Catalunya, where she developed interest in fields such as deep learning and reinforcement learning. She is currently an intern at HEIG-VD working on the SIA-REMU project, applying RL techniques to decide which ambulance to send in an emergency, this is done as part of her master thesis.
Enric is a computer scientist who graduated from Universitat Politècnica de Catalunya and is currently a student of a Master’s in Artificial Intelligence at the same university. He has always been interested in solving real-world problems through technology and informatics. Some examples of his work can be the construction of a therapeutic exoskeleton with Nitinol or the implementation of an intuitive and low cognitive load graphical interface to improve the interaction between humans and collaborative robots. During his formation and, more concretely, during his master, he developed an interest in Artificial Intelligence fields such as Computer Vision, Deep Learning and Reinforcement Learning. Now he is an intern working on an Offline Reinforcement Learning approach for SIA-REMU project as part of his master’s thesis.
Bachelor and Diploma students: link
Former members of our team
Kang graduated the Department of Physics in 2007 and received master’s degree in mechanical engineering in 2009 at KAIST. He received his Ph.D. in Physics from the Department of Physics, KAIST in 2017. After working in the same laboratory as a postdoc for about a year, he moved to Switzerland and worked in the Institute of materials, EPFL from 2018 to 2022. He has studied electromechanical phenomena at nanoscales, especially the coupling of electric and strain fields in the ferroelectric domains and their boundaries, by experiments, theory, and computational simulation.
Dr. Regis HOUSSOU
Senior Research Data Scientist, Ph.D in Financial Mathematics, University of Neuchatel, Switzerland. Master of Advanced Studies (MAS) in Quantitative Finance, ETH Zurich, Switzerland. BSc and MSC in Applied Mathematics, University of Neuchatel. Dr. Regis Houssou obtained his Ph.D. in Financial Mathematics. During his Ph.D, he developed and implemented new quantitative approaches for the pricing of defaultable bonds and credit derivatives. His research interests include Credit Risk Models, Option pricing, Stochastic Processes, Partial Differential Equations, Machine Learning and Statistical Analysis. Dr. Regis Houssou has several years of experience working; he has worked as Quantitative Analyst at Fundo SA in Lausanne, as Statistician at Swiss Institute of Bioinformatics (SIB) in Lausanne and as Data Scientist at the United Nations Migration Agency (IOM) in Geneva. He is currently a Data Scientist at HEIG-Vd, working in collaboration with NetGuardians.
Research Data Scientist. Cezar Harabula graduated from the French École Polytechnique (ingénieur) and holds a MSc from Eurecom Sophia Antipolis in computer science. He is a software engineer who worked in Silicon Valley and Paris (at Murex or in Astek’s incubator), built his startup, and later obtained a PhD in quantum devices from the University of Basel. The noise-based stochastical analyses he elaborated during his doctoral studies offered precise descriptions of novel electronic transport phenomena in quantum dots. After a deep-learning intermezzo at ETHZ, he integrated the fraud detection project we are running with NetGuardians.
Now Assistant Professor at the university of Leeds, UK. PhD in Mathematics, University of Lausanne. MSc in Probability Theory, Nankai University, China. BSc in Information and Computer Science, Agriculture University of Hebei, China. Dr. Ji received an award from the Fondation Nicolas and Helena Porphyrogenis (Lausanne) 2014 for his excellent doctoral thesis entitled « Ruin and related quantities in some advanced insurrance risk models ». His research interests include stochastic processes, probability theory and machine learning.
Now at PARC, Palo Alto, California, USA. MSc and PhD in Computer Science, TU Delft, The Netherlands. Dr. Alexander Feldman (Alex) is a visiting scientist at HEIG-VD (used to be a post-doc in the same institution). He has obtained his Ph.D. (cum laude) in computer science/artificial intelligence and M.Sc. (cum laude) in parallel and distributed systems from the Delft University of Technology, The Netherlands. He is a visiting researcher at the Delft University of Technology and PARC (former Xerox PARC). He has published in leading conference proceedings and international journals covering topics in artificial intelligence, model-based diagnosis, and engineering. In cooperation with NASA Ames Research Center and PARC, Alexander Feldman has co-organized the First and Second International Diagnostic Competitions. Alexander Feldman’s interest cover wide spectrum including topics such as automated problem solving, software and hardware design, design of diagnostic space applications, digital signal processing, and localization. Prior to the start of his academic career Feldman has several years of experience working for the Market Risk Management of ING Bank, The Netherlands and Zend Technologies, Israel.
Hugo graduated from the Blaise-Pascal University (currently Université d’Auvergne) in Clermont-Ferrand, France. He obtained his PhD degree in Theoretical Physics in 2012 which was awarded the best young researcher prize for his work on light-matter interaction in semiconductor nanostructures. Early 2013, he joined the group of Professor Vincenzo Savona at EPFL and worked as a Postoctoral Researcher/Senior Scientist for 5 years in the field of Quantum Photonics. He used to be a Ra&D Researcher in the group of Professor Stephan Robert at HEIG-VD. He used to be in charge of the Artificial Intelligence research effort in building predictive models for healthcare logistics in collaboration with Calyps Switzerland.
Antoine FRIANT, BS in Computer Science
Now at Calyps. Data Scientist. HEIG-Vd (BSc). He joined HEIG-VD in 2019 as a full time Machine Learning Engineer with the goal of optimizing logistics in hospitals thanks to deep learning models. Antoine holds a Bachelor’s degree in Software Engineering from HEIG-VD, where he worked on his Bachelor’s thesis : using machine learning to make population density maps from nighttime satellite imagery.
Former Master Students and Interns
Final year of BSc studies, IAESTE intern. During my studies, I developed an immense interest in machine learning and AI. I first started with computer vision, as it was easy for an inexperienced student such as myself to visualize, learn, and apply knowledge in such an intuitive field. However, with growing experience, my interest shifted to NLP and RL. Up until now, I have only practiced exercise tasks in those fields, so I am very excited to work on the SIA-REMU project, which will be my first contribution to a real research task. In the future, I am planning to obtain a PhD degree, and hope to one day lead a research team and lecture as a university professor. Beside my chosen field of study, I am interested in playing music, learning languages, snowboarding, traveling and exploring urban areas. Now at the University of Tokyo.
Master student in Mathematics, IAESTE Intern. He graduated from the TU Vienna in Technical Mathematics (BSc). From 2020 to 2021, he did an internship at Amundi Austria GmbH, where he was experimenting with different dimension-reducing methods (such as PCA, t-SNE) on high-dimensional financial data. Currently, he is an Intern at HEIG-VD, working on the SIA-REMU project (https://www.interreg-francesuisse.eu/beneficiaire/sia-remu/).
Master student in Data Science at the Autonomous University of Madrid (UAM). Graduated in physics at the Autonomous University of Madrid in 2021. In the last year of my Bachelors Degree I took part in a project that consisted on using Genetic Algorithms to fit the Sound Horizon as a function of different cosmological parameters,
culminating on the publishing of a paper on Physical Reviews D . Now I am working as an intern at HEIG-VD as part of the project SIA/REMU which aims to make emergency management more efficient.
Francisco José SIMORE
A final year MSc. in Industrial Engineering student in Argentina. During his studies, he got a part-time research internship scholarship at the « National Scientific and Technical Research Council » with the objective of optimizing public transport. He currently is an intern at HEIG.VD and is part of a project which focuses on choosing the most appropriate vector in face of a medical emergency and also its optimal location. He has a keen interest in data science, operation research, and software development.
Computer Security, IAESTE Intern (Fall 2019-20). I’ve graduated from Glasgow Caledonian University in Digital Security, Forensics and Ethical Hacking (BEng) with a first-class honour. While undertaking my studies, I became familiar with the various aspects that make up communications and their inherent threats. I am very interested in machine learning and in its application in the security landscape and I was glad to research this aspect as I was carrying out my thesis about the detection of malicious emails. I was an intern at HEIG-VD through an IAESTE placement.
Now at GeorgiaTech, USA. Ph.D, Seoul National University, Korea. There are two his research topics: (Internet) Data Science, and Future Internet Technologies. For the data scientist, he focuses on the users’ behaivoral characteristics and content propagation, which published several papers in ACM SIGMETRICS, WWW, and ACM COSN. In the master period, he designed and implemented a Content-Centric Network (CCN) router, which efficiently delivers contents to clients by removing redundant packets in Internet. Currently, he focuses on understanding how users consume online contents and various type of applications based on user consumption patterns. See more details in here (http://mmlab.snu.ac.kr/~djchoi)
Master student in Scientific Computing and Mathematics of Information (MSc) at the University of Strasbourg (Spring 2020). This degree is based on PDEs, control theory, signal processing and deep learning. I’ve graduated from the University of Strasbourg in Pure Mathematics (BSc) in 2018. Since October 2019, I am working on a project with Electricité de France R&D PERICLES to design a deep learning model to solve fluid dynamics problems. (Navier-Stokes PDEs). From May to August 2019, I was an intern at Cross Valorem Luxembourg and I worked on financial models. I also worked on an NLP problem for Electricité de Strasbourg in 2019. My internship at the HEIG-VD starting in January 2020 will be based on setting up an AI-based algorithm for banking fraud detection. »
Master student in Mathematics, IAESTE Intern, Gdansk University (Spring 2020). graduated in the field of Financial Mathematics at Gdansk University of Technology in July 2017 and since then have been working for PwC Poland, Financial Crime Unit. During my first year at the company I was working in the operations department as an AML Analyst. I decided to take one year off before my master degree because of the opportunity to work for such a well-known company straight after my bachelor studies. A few months later, I took part in a project at the University San Francisco de Quito in Ecuador. During my internship in Ecuador I was working on an optimization model for a company delivering liquids. When I came back from Ecuador I changed my career path to join the Technology Stream where I took on the role as a Data Analyst. Since October 2018 I have been studying Big Data at the University of Gdansk. During the summer of 2019 I was enrolled at the University of Macau in Macau, China, Department of Computer & information Science where I was involved in a project based on complex functions visualization using the domain coloring method.
Sahni is an associate professor of marketing at Stanford GSB. His research investigates consumer and seller decisions in digital marketplaces. His research employs methodologies ranging from econometric analysis to large-scale experimentation to make inferences from data. In recent years, he has studied how platform policies can increase welfare of all sides of a marketplace. His research also examines the effects of various marketing activities including internet search advertising, display advertising, retargeting, and email marketing. He has published numerous articles in leading marketing and economics journals, including Marketing Science, Management Science, Quantitative Marketing and Economics, and the Review of Economic Studies. His research has been recognized with awards, including American Marketing Association’s Paul Green Award, and Quantitative Marketing and Economics Journal’s Dick Wittink prize. He was named to the Marketing Science Institute’s list of Young Scholars in 2017.