In the near future, the "superorganism mankind" will be confronted with an avalanche of drastic changes, so-called "game changers". All living beings and the communities they form are complex interaction networks. They are made up of a multitude of components that influence each other and are therefore difficult to understand and predict. Only with a comprehensive understanding of the complex systems of our living environment can society be prepared for these changes. The understanding of such systems should provide science, business and politics with the necessary basis for decisions in terms of sustainability.
COLIBRI's complexity research of living systems stands, respectively concentrates on three pillars, namely “biological sciences”, “social sciences” und “computational complexity sciences”. All scientists participating in COLIBRI are working on complex systems in their respective fields. They use analytical as well as computer-based methods of strategic interaction theory (game theory), network analysis, optimization, and reconstruction. In the following, topics that are investigated in the three research pillars are presented:
The members of our group share a very focused set of central research questions:
- Identifying and modeling of complexity and emergent phenomena of self-organization across multiple disciplines.
- Transfer methodology to unveil this complexity across disciplines.
- Application of this knowledge in various practical forms, ranging from technical applications to policies.
- Investigating enhanced homeostasis by diversity.
- How does entropy decrease in open systems?
- Studying systems far from equilibrium.
- What are the underlying principles of complex systems?
- What are critical parameters of such systems’ sustainability (e.g., stability, robustness and resilience), of their propagation, and of their change?
- What effects have interferences with other complex systems, and how can complex systems be managed/controlled/influenced/abused?
Ultimately, a rich set of specific and practical questions can be addressed with this knowledge: e.g. on how can we efficiently and reliably communicate with swarms of animals/agents to control them? How do invasive plants restructure the previous vegetation patterns? How do fake news spread in (social) media systems to control the public mindset? ... and many more problem sets of this kind. Gathering knowledge on these and similar high impact issues is essential to address pressing problems of our times.
Thereby key aspects of our research are:
COMPLEX GRAPHS and NETWORKS: We study complex transport networks and complex chemical reaction networks, incl. their coordination, self-organization and optimization. Ecosystems are inherently networks and evolutionary dynamics inherently produce tree structures (tree of life). Social systems and markets are dynamically reconfiguring networks that are not intuitively predictable due to their non-linear and heterogeneous network characteristics.
COMPLEX DECISIONS: Clearly, decisions become more complex if they involve the inclusion of several stakeholders with more or less diverging views. These collective decision-making processes reach from consensus building in small committees to popular elections of millions of voters. Algorithmic Decision Theory investigates properties and limitations of solution concepts in this area. Other decisions concern in particular the allocation of different resources as studied in the research area of “fair division”.
COMPLEXITY of EVOLUTION: Evolutionary game theory in economics and biological evolution; Evolution is driven by ecology and ecology is created by evolution. The ever-branching “tree of life” and the complexifying network of ecological relationships are tightly intertwined and, to understand either one of those issues, one has to understand the other. In a similar way, market structures, social fragmentation and market dynamics and social dynamics are intertwined and, for a full understanding, have to be tackled by a holistic approach, in order to de-complexify the system.
COMPLEXITY of ALGORITHMS: The theoretical study of algorithms involves the precise classification in the hierarchy of computational complexity, the determination of approximability or intractability status, and the search for alternative solution concepts. This applies for all algorithmic aspects that arise in the above- mentioned network problems, in algorithmic decision theory, and in the broader field of complex modelling.
COMPLEX NON-LINEAR SYSTEMS: We research tipping point prediction and study emergent phenomena. These phenomena are inherent to complex systems across biochemistry, biology, economics, logistics, traffic, societies and we approach them with a shared pool of common methodology.
Our joint methodologies are: Computer simulation, mathematical modeling, formal methods, strong algorithmics and agent-based models represent our main shared methodology pool to be taught coordinately to our students.