We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Condensed Matter > Statistical Mechanics

Title: Mean-performance of Sharp Restart II: Inequality Roadmap

Abstract: Restarting a deterministic process always impedes its completion. However, it is known that restarting a random process can also lead to an opposite outcome -- expediting completion. Hence, the effect of restart is contingent on the underlying statistical heterogeneity of the process' completion times. To quantify this heterogeneity we bring a novel approach to restart: the methodology of inequality indices, which is widely applied in economics and in the social sciences to measure income and wealth disparity. Using this approach we establish an `inequality roadmap' for the mean-performance of sharp restart: a whole new set of universal inequality criteria that determine when restart with sharp timers (i.e. with fixed deterministic timers) decreases/increases mean completion. The criteria are based on a host of inequality indices including Bonferroni, Gini, Pietra, and other Lorenz-curve indices; each index captures a different angle of the restart-inequality interplay. Utilizing the fact that sharp restart can match the mean-performance of any general restart protocol, we prove -- with unprecedented precision and resolution -- the validity of the following statement: restart impedes/expedites mean completion when the underlying statistical heterogeneity is low/high.
Subjects: Statistical Mechanics (cond-mat.stat-mech); Probability (math.PR)
DOI: 10.1088/1751-8121/ac16c5
Cite as: arXiv:2102.13154 [cond-mat.stat-mech]
  (or arXiv:2102.13154v1 [cond-mat.stat-mech] for this version)

Submission history

From: Shlomi Reuveni [view email]
[v1] Thu, 25 Feb 2021 20:20:23 GMT (290kb,D)

Link back to: arXiv, form interface, contact.