CORE group investigates methods and concepts to assess the stability of highly dynamic systems and define strategies to manage the underlying complexity.
Understanding and modelling natural and artificial systems is notoriously hard due to their inherent complexity. This complexity is a result of the diversity and opacity of interactions both within and outside these systems that lead to properties like non-linearity, uncertainty, or spontaneous transformations. Without the ability to efficiently monitor, assess, and predict the behavior of such systems, it is increasingly challenging to anticipate adverse incidents and manage their (negative) effects. Therefore, the primary objective of the CORE group is to develop novel analytical approaches for assessing the complexity and resilience characteristics and developing strategies to enhance their performance. For this, the group combines algorithmic and formal methods from complexity science with Artificial Intelligence to build better models that represent real-world systems and create the ability to manage their properties.
Within a highly interdisciplinary team, our research topics include (but are not limited to):
Techniques for advancing machine learning capabilities, including preprocessing, processing, and post-processing.
Enhancing transparency, interpretability, and explainability of Artificial Intelligence and Digital Twins
Studying the underlying rules of natural and man-made systems to understand behaviors ranging from nonlinearity to multilayered interactions between different components. By deducing mechanisms that regulate biological systems, we provide new insights into assessing software aging and maintenance.
Investigating and advancing methodological approaches in Information Theory and Complexity Science for multivariate time series analysis and complex matrices. The goal is to gain insights into the dynamics of various systems and to study underlying patterns for identifying errors, security risks, early warning signals, and enhance prediction accuracy of Machine Learning tasks.
Exploring quantum-inspired machine learning methods for identifying abnormal system behavior in network structures and reducing functional risk.
The Complexity and Resilience Group participates in the following research projects:
