Upon completion of this course, the students will gain a comprehensive overview of the theoretical foundations of swarm intelligence as well as the practical features for developing specific applications.

After successful participation in this course students shall be able to:

• analyze, disseminate and communicate the topic issues, concepts and problem definitions presented in class;
• understand the basic principles of swarm intelligence;
• model intelligent social agents in complex landscapes, map and adapt key principles of swarm intelligence to engineering applications;
• use up-to-date efficient variants of different swarm-based metaheuristics (together with their potential applications), analyze them and prove their properties;
• evaluate the power and limitation of swarm intelligence when it comes to solving problems;
• develop simulation models based on swarms of intelligent agents.

Swarm intelligence deals with natural and artificial systems composed of many individuals that coordinate using decentralized control and self-organization, which consequently form an emergent behavior. In particular, the discipline focuses on the collective behaviors that result from the local interactions of the individuals with each other and with their environment. Such systems are highly adaptable, robust, flexible, and capable to function autonomously. The swarm-based mechanisms can be very useful in technical systems as well as in optimization of complex problems.

We focus especially on the concepts of swarm evolution, collective intelligence, stygmergic communication and self-organization, and discuss different swarm intelligence algorithms that belong to the set of probabilistic metaheuristic approaches.

During the course, a number of swarm intelligence systems (bee colonies, ant colonies, schools of fish, flocks of birds, herds of land animals) will be presented together with the corresponding swarm intelligence algorithms that will be analyzed and compared. Emphasis is given to such topics as the modeling and theoretical analysis as well as the real applications from diverse domains to show the usefulness of this approach. The other “accompanied” aspects such as parameters setting and tuning will be also presented.

The following topics will be covered:

Part 1: Fundamentals of swarm intelligence (stability analysis, swarm aggregation, swarm in known and unknown environments, dynamic optimization);

Part 2: Swarm-based metaheuristics (theoretical foundations, swarm clustering and sorting, Particle Swarms, Ant Colony, Artificial Bees, Fireflies Algorithm, Bacterial Foraging, recent advances and new inspirations);

Part 3: Applications (swarm robotics, internet computing, software engineering, sensors, data mining,…) regarding different optimization problems, finding optimal routes, scheduling, routing, structural optimization, image and data analysis.

Literature

• Eric Bonabeau, Marco Dorigo and Guy Theraulaz, Swarm Intelligence: From Natural to Artificial Systems, Oxford University Press, 1999
• James Kennedy and Russel Eberhart, Swarm Intelligence, Morgan Kaufmann, 2001
• Marco Dorigo and Thomas Stützle, Ant Colony Optimization, The MIT Press, 2004
• Andries Engelbrecht, Fundamentals of Computational Swarm Intelligence, Wiley 2006
• Veysel Gazi and Kevin M. Passino, Swarm Stability and Optimization, Springer, 2011
• Aboul Ella Hassanien and Eid Emary, Swarm Intelligence: Principles, Advances, and Applications, CRC Press, Taylor & Francis, 2015

Assessment

• Written exam (60%)
• Assignments (Project) (40%)

The student will pick a project topic among the proposed assignments. Following deliverables are required: the code implemented and a short documentation (4 to 6 pages). Further, the student will present his/her work in an interview.

The course prerequisites include programming skills and fundamental knowledge of agent-based modeling, probability calculus and analysis (differential equations, continuous and discrete time).