(658c) Emergence of Structural Features in Complex Networks Via Adaptation

Chemical and biological systems can often be represented in the form of networks which show interactions between the various variables involved. Understanding the complexity of these systems is crucial for any systems engineering application. One such measure to understand network complexity is entropy. Venkatasubramanian et al. [2006] have suggested entropy maximization as a general design principle for networked systems. More recently, Stevens [2016] observed that biological systems seem to follow a maximum entropic distribution. From a goal-driven setting, a system’s objective is survival, which consists of a short-term component and a long-term component. Venkatasubramanian et al. [2004] suggested that the short-term and the long-term components can be represented using the efficiency of the networked system (to reach a goal) and robustness of the system to external disturbances (such as, node-deletion), respectively. In this work, we explore these design possibilities for various chemical and biological systems. We also study the correlation between the network topology exhibited by these systems and their robustness/efficiency metrics. This motivates us towards an information theory-based measure of complexity of networked systems. How this information theoretic measure relates to the systems’ robustness and efficiency, is studied as well.

References

Charles F Stevens. A statistical property of fly odor responses is conserved across odors. Proceedings of the National Academy of Sciences, 113(24):6737–6742, 2016.

Venkat Venkatasubramanian, Santhoji Katare, Priyan R Patkar, and Fang-ping Mu. Spontaneous emergence of complex optimal networks through evolutionary adaptation. Computers & chemical engineering, 28(9): 1789–1798, 2004.

Venkat Venkatasubramanian, Dimitris N Politis, and Priyan R Patkar. Entropy maximization as a holistic design principle for complex optimal networks. AIChE journal, 52(3):1004–1009, 2006.