Conquering data series: Computer scientist from Kaiserslautern receives two million euros from the European Research Council

The European Research Council (ERC) has awarded Professor Dr. Anthony W. Lin an ERC Consolidator Grant, one of the most prestigious research grants for European researchers. The computer scientist from Kaiserslautern will receive around two million euros for five years to develop new methods for analysing complex computer programs and proving that they are error-free. One core aspect of such programs is the presence of data series. Lin's team is initially developing theoretical models that they want to put into practice.
 
Whether at health insurance companies, online shops, in the private cloud, on the production line in an industrial company or at listed corporations - nowadays, huge amounts of data accumulate in a wide variety of areas. A large amount of such data are represented as data sequences, for example as a sequence of numbers. "Modern programs constantly process data sequences, for example data structures such as arrays, which computer science students learn about in computer science 101, or the price history of a stock. Due to the complexity of such programs, errors easily creep in. The challenge is to develop new methods that lead to both error-free and efficient programs while making programming easier," says Professor Dr. Anthony Lin, who conducts research on "Automated Reasoning" at the University of Kaiserslautern-Landau.  
 
But how can one guarantee that such a computer program is error-free? That is what the ERC-funded project "Logic and Automata over Sequences with Data" (LASD) will address. The solution lies in the development of a new automata theory.
 
Among other things, Lin's team is looking at how so-called graph databases can be queried more easily and efficiently. These can be used to store highly interconnected information. They are used, for example, to record financial transactions between several parties. They contain a variety of data on account holders (individuals or companies) and transactions, such as names, ages, amounts of transfers, etc. In order to detect fraud in such systems, the traffic of transfers, i.e. the sequences of transfers, as well as their amounts are of particular interest. Up to now, it has been very time-consuming to filter out this information.

Another area of application that will be the focus of the project is data series in which changes occur over time that cannot be foreseen in advance. In the field of machine learning, various methods have recently been developed that receive a data series as input and make a decision. For example, such systems are used on loan applications, to decide whether an application is to be approved or not. "The main problem is that the decision of such a system lacks explanation and transparency, for example, why should the application for a loan be rejected?" continues Professor Lin. "This explainability problem is a key issue in the field of machine learning. We want to develop a theory that could help provide more explainability." 

Professor Lin has been conducting research as a professor in "Automated Reasoning" at University Kaiserslautern-Landau since 2019, as well as a Max Planck Fellow at the Max Planck Institute for Software Systems. The Max Planck fellowship program promotes collaboration between outstanding university professors and scientists from the Max Planck Society. Previously, Professor Lin was at the University of Oxford in England, where he has already received an ERC Starting Grant. 

There are currently three other researchers at University Kaiserslautern-Landau whose research is funded by the ERC: Professor Johannes Herrmann from the Department of Biology has received an ERC Advanced Grant to study how cells remain functional over time. Professor Mathias Weiler from Physics was awarded an ERC Consolidator Grant. He and his team are working on spin waves and new spintronic devices that could significantly speed up the storage, processing and transmission of information. His physicist colleague, junior professor Phillipp Pirro, is building an artificial brain as part of an ERC Starting Grant, in which information is to be transmitted with the help of magnons, the quantum particles of spin waves.

Contact  
Professor Dr. Anthony W. Lin
Automated Reasoning Group, Department of Computer Science
University Kaiserslautern-Landau
Phone: 0631 205-3330
E-mail: lin[at]cs.uni-kl.de