ABSTRACT
This chapter focuses on an additional powerful principle that allows both engineers and biological evolution to scale up and produce remarkably complex systems that work in the real world: modularity. The structural modularity of a network can be quantified by algorithms, such as the Newmann–Girvan algorithm. These algorithms attempt to partition the network into parts, such that the parts have more connections inside them than to other parts. The fundamental reason for the lack of modularity in the evolved networks is that modular structures are far rarer and usually less optimal than non-modular ones for a given task. The chapter explores a mechanism for the evolution of modularity, using simulated evolution of circuits made of logic gates. These simulations will serve as a metaphor for understanding biological evolution. The chapter also presents some closing thoughts on the key concepts discussed in the preceding chapters of this book.
