Drift-diffusive resetting search process with stochastic returns: Speedup beyond optimal instantaneous return

Stochastic resetting has recently emerged as a proficient strategy to reduce the completion time for a broad class of first-passage processes. In the canonical setup, one intermittently resets a given system to its initial configuration only to start afresh and continue evolving in time until the target goal is met. This is, however, an instantaneous process and thus less feasible for any practical purposes. A crucial generalization in this regard is to consider a finite-time return process which has significant ramifications to the firstpassage properties. Intriguingly, it has recently been shown that for diffusive search processes, returning in finite but stochastic time can gain significant speedup over the instantaneous resetting process. Unlike diffusion which has a diverging mean completion time, in this paper, we ask whether this phenomena can also be observed for a first-passage process with finite mean completion time. To this end, we explore the setup of a classical drift-diffusive search process in one dimension with stochastic resetting and further assume that the return phase is modulated by a potential U(x)=λ|x| with λ>0. For this process, we compute the mean first-passage time exactly and underpin its characteristics with respect to the resetting rate and potential strength. We find a unified phase space that allows us to explore and identify the system parameter regions where stochastic return supersedes over both the underlying process and the process under instantaneous resetting. Furthermore and quite interestingly, we find that for a range of parameters the mean completion time under stochastic return protocol can be reduced further than the optimally restarted instantaneous processes. We thus believe that resetting with stochastic returns can serve as a better optimization strategy owing to its dominance over classical first passage under resetting.