The evolutionary pressure towards hierarchical networks

The Evolutionary Origins of Hierarchy: Evolution with performance-only selection results in non-hierarchical and non-modular networks, which take longer to adapt to new environments. However, evolving networks with a connection cost creates hierarchical and functionally modular networks that can solve the overall problem by recursively solving its sub-problems. These networks also adapt to new environments faster. Credit: Henok Mengistu et al./PLOS Comp. Bio

Telecommunications networks are hierarchical and the reason is simple: if you were to design a flat network where every phone is directly connected with all other phones the quantity of wires will skyrocket and become unmanageable (the number of connections will be n(n-1)/2  which means, given the present number of fixed line connections in the world 1,664,100,000,000,000,000 lines!). The cost and the practicality of deploying and managing such an astronomic number are just impossible.

According to an article published on PLOS Computational Biology Nature evolutionary process ended up in creating hierarchical structures for the same reasons.

Simulating in computer the evolutionary process through random changes that are selectively filtered based on the "cost" one ends up with modular and hierarchical structures. That would not be the case if the "cost" parameter is removed from the selection process.

Cost in "Nature" refers to the use of energy. The more energy is needed the higher the cost.

Our brain is organised in hierarchical, and modular, structure. The modularity leverages on the experience and replication of what works cutting the cost of trying new solutions. The hierarchy cuts the operation cost of the system.

Researchers, from the University of Wyoming and INRIA -an EIT Digital partner- in The Evolutionary Origin of Hierarchies are shedding light on this process and are providing hints on how to design software systems, particularly in the area of Artificial Intelligence, that could leverage on this discovery.  As we are moving towards software that can dynamically adapt itself, introducing in the adaptation mechanism the parameter of "cost" can lead to smarter algorithms. 

The field of application is quite large since in principle any system that is sufficiently complex can benefit from this approach, including Smart Cities.

Author - Roberto Saracco

© 2010-2018 EIT Digital IVZW. All rights reserved. Legal notice