The dynamics of networks enables the function of a variety of systems we rely on every day, from gene regulation and metabolism in the cell to the distribution of electric power and communication of information. Understanding, steering and predicting the function of interacting nonlinear dynamical systems, in particular if they are externally driven out of equilibrium, relies on obtaining and evaluating suitable models, posing at least two major challenges. First, how can we extract key structural system features of networks if only time series data provide information about the dynamics of (some) units? Second, how can we characterize nonlinear responses of nonlinear multi-dimensional systems externally driven by fluctuations, and consequently, predict tipping points at which normal operational states may be lost? Here we report recent progress on nonlinear response theory extended to predict tipping points and on model-free inference of network structural features from observed dynamics.

When it comes to integrating digital technologies into the classroom in higher education, many teachers face similar challenges. Nevertheless, it is difficult for teachers to share experiences because it is usually not possible to transfer successful teaching scenarios directly from one area to another, as subject-specific characteristics make it difficult to reuse them. To address this problem, instructional scenarios can be described as patterns that have been used previously in educational contexts. Patterns can capture proven teaching strategies and describe instructional scenarios in a consistent structure that can be reused. Because priorities for content, methods, and tools are different in each domain, a consensus-tested taxonomy was first developed with the goal of modeling a domain-independent database to collect digital instructional practices. In addition, this presentation will present preliminary insights into a data-driven approach to identifying effective instructional practices from interdisciplinary data as patterns. A web-based application will be developed for this that can both collect teaching/learning scenarios and individually extract scenarios from patterns for a learning platform.

Computer vision and AI methods are percolating many branches nowadays. Also in the research field of real estate rating computer vision and AI methods have lead to very interesting innovations. In this research talk, real estate classification by AI-enabled computer vision techniques is discussed. The talk will give an overview of recent research efforts in the field and focus on latest findings of our research group. This consists of age or heating demand prediction of real estates by photographs as well as the analysis of satellite images for detecting building footprints.

IoT-enabled applications increase tremendously in various sectors, such as transportation, healthcare, education, agriculture, and so forth. These applications sense properties using sensors, perform intelligence, and apply the findings using actuators. Instead of submitting sensor data directly to the cloud, intelligence could be performed with the inclusion of several edge/fog nodes. This improves the privacy and computation time of applications. This talk will provide insights on edge intelligence techniques for such IoT-enabled applications. In addition, a few protocols that are involved in such applications are discussed.