Forskargrupp We are interested in agents that reason, and in reasoning about agents. We focus on studying reasoning principles, and on making software systems reason better.
Software plays a crucial role in running our modern society, often as integral part of complex information systems that help orchestrate and coordinate how organizations work. Examples are enterprise resource planning systems that manage functions such as logistics and procurement in companies, and public administration systems that process citizens' tax returns. These systems are notoriously complex, often integrating many subsystems across several organizations. Most subsystems evolve over a long period of time (typically several decades), leading to extreme heterogeneity in terms of technologies that are, however, tightly coupled to each other. This makes it difficult to understand the systems' inner workings, i.e., the why and how behind specific decisions as well as broader behaviors a system gives rise to.
The practical objective of our research is to improve how we can reason about such complex systems, such that we get better at improving the systems themselves, to make them ultimately better at serving our society. For this, we take an agent-based view on complex systems, as well as a principle-based approach to understanding fundamental reasoning behaviors.
In our research, we frequently study fundamental challenges of principle-based reasoning, primarily when it comes to reasoning using graphs that model mutually conflicting or supporting statements (formal argumentation). For example, have taken principles from economic theory and applied them to graph-based reasoning, and we have formalized well-known reasoning fallacies to show that they occur in fundamental methods of reasoning.
On the applied side, we research reasoning about agents and multi-agent systems, as well as inference in software systems that form the backbone of large organizations. We frequently collaborate with leading technology companies, such as SAP and Ericsson. Some of our work has informed product visions and roadmaps in the software industry and made it into production-grade software systems.
The group collaborates with many other researchers across the world, for example with groups at Imperial College London, the Technical University of Munich, the University of Vienna, and the University of Melbourne.
Our research is mainly supported by the Wallenberg AI, Autonomous Systems and Software Programme, WASP, funded by the Knut and Alice Wallenberg Foundation.
