Theory of Swarm Algorithms and Their Effectiveness in Uncertain Environments (TOSU)
Summary: Bio-inspired swarm algorithms like ant colony optimization are well established in practice for solving optimization problems with complex constraints even in difficult domains involving uncertainty. The goal of this project is to further the understanding of such swarm-based search heuristics by making theoretical analyses. In particular, we want to make major contributions regarding dynamically changing and stochastic objectives, for which swarm algorithms have been empirically observed to be successful, but theoretical analyses are just starting. Furthermore, we want to continue our work on the foundations of swarm intelligence by proving tight runtime and quality bounds for elementary static problems, for which new techniques seem to be required.