The research activities in the CAOS group are varied, unified by a general interest in the use of bottom-up techniques in:
- the design and optimization of efficient technological systems, processes, products, tools, and art;
- in the investigation of emergent (adaptive, self-organising) phenomena in social, biological, and technological systems.
The questions we want to answer are (1) how interactions give rise to patterns of behavior, (2) how can complex systems be described, and (3) how are complex systems emerging through pattern formation and evolution. The used methods include cellular automata, networks, and agent-based models.
Natural evolution has been creating complex biological organisms for an unimaginable number of years. Its ability to create complex designs and solutions that evolve and adapt to changing environments have inspired researcher to apply biologically-inspired methods to artificial systems. The used methods include evolutionary computation, artificial development, morphogenesis, swarm intelligence.
Research by Design
Innovative, unconventional, complex designs are usually the result of an iterative and creative research process, where components are synthesized bottom-up rather than analyzed top-down. The used methods are design thinking, user-centered design, participatory design, evolutionary design.
Statistics and Machine Learning
Complex real life problems and systems often produce big amounts of data. Data analysis is carried out through statistical and mathematical models and tools. The used methods are statistics, game theory, and machine learning techniques.
Artificial Intelligence and Robotics
Many of the systems that surround us incorporate some form of intelligent or cognitive abilities. Such properties are not easy to achieve in artificial systems, to the same degree they appear in natural and biological systems. How does intelligence emerge? What are the necessary properties for cognition? How can such properties be replicated in machines? The applied methods are: neural networks and deep learning, recurrent and convolutional networks, neuro-evolution, evolutionary and swarm robotics.
Artificial Life and Living Technology
What are the essential properties of life organisms? And how can we recreate some of the life-like behaviors in artificial systems, e.g., growth, reproduction, adaption, intelligence? How can communities of biological and artificial systems co-exist? Artificial life (Alife) deals with the study or life-as-it-might-be rather than life-as-we-know-it. The used methods include soft-, hard-, and wet- alife.
The mandate of the CAOS Lighthouse initiative is to create a real multi-disciplinary and cross-departmental collaboration at OsloMet-TKD, together with national and international partners. The long-term goals are:
- Establish a long-term collaboration around the themes of Complex Systems and Applied Artificial Intelligence;
- Establish a multidisciplinary master specialization where the required methods are taught;
- Establish a multidisciplinary lab where students can apply such methods
- Support ongoing and future research projects;
- Run a periodic workshop in collaboration with other national research environments;
- Identify synergies with other initiatives/labs at OsloMet, and in other national and international institutions, establish real collaboration.