Title:Emergence of Lévy walks from second order stochastic optimization

Abstract:Many living organisms seem to perform two different types of search: directed search (taxis) and random search. The former is observed when the environment provides enough cues to guide the motion towards the target. The latter may not involve any memory and information processing and is mathematically modeled by random walks. We show that both types of motion can be generated by a common mechanism in which Lévy flights or Lévy walks emerge from a second-order gradient-based search with noisy observations. No explicit switching mechanism is needed -- instead, a continuous transition between the directed and random motions emerge depending on the Hessian matrix of the cost function. For a wide range of scenarios the Lévy tail index is given by $\alpha=1$, in line with previous experimental observations in foraging organisms. Our results suggest that adopting a second-order optimization method can be a useful strategy to combine efficient features of directed and random search.