Random walks on networks with stochastic resetting
Denis Boyer (UNAM, Mexico)
Resetting a stochastic process from time to time to the initial state can represent an efficient search strategy for locating a hidden target. This fact finds applications in statistical physics, computer science, enzymatic reactions or foraging ecology. Random walks on general networks and subject to resetting have not been studied, despite their relevance to epidemic spreading, searching on the web or human mobility, among others. We develop an extension to an arbitrary network topology of the diffusion problem with stochastic resetting. Our formalism applies to finite undirected networks and is implemented from the spectral properties of the random walk without resetting. We show that resetting can allow a faster exploration of many network architectures such as rings, Cayley trees, random networks and complex networks, highlighting the importance of the choice of the starting/resetting node. The small-world effect and the presence of communities, two properties of many real world etworks, strongly affects exploration strategies based on resetting.