Optimal control of aging in complex networks

Aging is a shared process of biological and technical systems. As structural and organizational complexity increases, the phenomenon of aging--the progressive increase in the probability of death or decay--arises as an emergent property. A key question is how to maximize longevity of an aging system at minimal cost of maintenance and intervention. Here, we answer this question using optimal control theory and machine learning on a network model of aging. We derive and numerically validate optimal protocols for repair that emerge from a balance between maximizing system healthspan and minimizing the overall cost of repair. These protocols may motivate the design of rational strategies for delaying aging in complex systems.