XMANAI partner spotlight - AiDEAS
Name: Dr Serafeim Moustakidis
Job title: Co-founder / CTO
Bio: Dr Serafeim Moustakidis has wide experience in computational intelligence, machine learning and data processing with more than 12 years of research experience in various fields. His research focuses in developing novel algorithms for solving important existing and/or emerging problems. His main scientific interests cover various application fields such as Deep Learning, Big Data, Biomechanics, Bio-economy, Health, Remote Sensing, Energy Optimization, Non- Destructive Testing (NDT) and machine learning-empowered imaging. He has been involved in the technical implementation, scientific or overall management of 25 R&D projects of a total budget of 35 million Euros. Serafeim has also been active in EU proposal writing since 2010 and has managed to secure more than €9.2 million (in FP7 and Horizon2020). He has worked for several research organizations across Europe.
Name: Patrik Karlsson
Job title: Co-founder / CTO
Bio: Patrik holds a MSc. in Materials Science from Uni of Nancy. Has worked for several years at Fraunhofer Institute on Electron beam/Plasma techniques. He has been involved in numerous Industrial Research projects and more than 20 years in EC funded projects. Has participated and lead more than 32 Research & Innovation Projects (being the coordinator of six) of a total budget greater than €105 million. Research involved nanotech&advanced materials, eLearning, AAL&rehabilitation, wearable sensors, WSN for agriculture, robotics & mechatronics, NDT, machine learning for defect detection and classification. In addition to his strong PM and technical background he also has significant financial and exploitation expertise.
Name: Lia Dikopoulou
Job title: Senior Data Analyst
Bio: Dr. Lia Dikopoulou was awarded with the academic degree of Doctor of Sciences: Computer Science from Hasselt University, Belgium. She is the author and co-author of scientific published papers, books and book chapters. In addition, she is a developer of the ‘fcm’ package in the R programming language. She has research experience working at various national and European projects and she currently works as a Senior data analyst at the AiDEAS company, under the H2020 EU XMANAI project. Finally, her research interests are focused on probabilistic graphical models, machine learning, graph theory, fuzzy cognitive maps, decision support systems and aggregation methods.
Q: What is your organization’s role in XMANAI?
A: One of the main duties of AiDEAS is to gather and analyze statistical data and transform them into useful information that can be used for critical decision-making. Moreover, to derive expert knowledge from our end users, employ AI methods to represent graphically the obtained knowledge and deliver final simulation reports for the partners, enabling them to take important decisions based on various facts and trends. Specifically, AIDEAS is leading Task 4.3 (Cross-Validation and Experts Evaluation of XMANAI AI Models), whereas it also significantly contributes to the development of XAI solutions that analyze and interpret patterns and trends in complex metrological data sets that could be helpful for optimizing UNIMETRIK’s scanning processes.
Q: How does XMANAI relate to your team’s background and interests?
A: AiDEAS staff comes from different computer science backgrounds with over 15 years experience and a particular focus in developing novel trustworthy algorithms as well as leading award-winning academic research for solving important existing and emerging problems in various industries such as Healthcare, Industry 4.0, and Oil & Gas. The use of explainable AI is within the company’s core expertise and interests.
Q: What is the novelty of XMANAI and the main benefits envisaged for your organization?
A: It is well-known that the main disadvantage of the machine learning systems is the lack of explainability. That’s why they are called black-boxes! XMANAI project will provide innovative methods to explain to stakeholders why certain predictions/decisions are made. Focusing on the transparency of AI processes, the company will produce more explainable models while maintaining a high level of learning performance/prediction accuracy, people will understand, trust, and effectively manage the results of AI technology. Already within the first twelve month of the project, we have developed a novel explainability tool which is called gLIME as a combination of graphical LASSO and LIME methods.
Q: Which target groups can benefit from XMANAI?
A: We expect that the user-friendly environment of the XMANAI platform will assist not only Data Scientists but also Business-users with or without programming skills to develop interpretable decision-making models from their data.