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


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


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

Computer Science > Computer Science and Game Theory

Title: Random Assignment Under Bi-Valued Utilities: Analyzing Hylland-Zeckhauser, Nash-Bargaining, and other Rules

Abstract: The Hylland-Zeckhauser (HZ) rule is a well-known rule for random assignment of items. The complexity of the rule has received renewed interest recently with Vazirani and Yannakakis (2020) proposing a strongly polynomial-time algorithm for the rule under bi-valued utilities, and making several general insights. We study the rule under the case of agents having bi-valued utilities. We point out several characterizations of the HZ rule, drawing clearer relations with several well-known rules in the literature. As a consequence, we point out alternative strongly polynomial-time algorithms for the HZ solution. We also give reductions from computing the HZ solution to computing well-known solutions based on leximin or Nash social welfare. An interesting contrast is that the HZ rule is group-strategyproof whereas the unconstrained competitive equilibrium with equal incomes rule is not even strategyproof. We clarify which results change when moving from 1-0 utilities to the more general bi-valued utilities. Finally, we prove that the closely related Nash bargaining solution violates envy-freeness and strategyproofness even under 1-0 utilities.
Subjects: Computer Science and Game Theory (cs.GT); Data Structures and Algorithms (cs.DS)
MSC classes: 91A12, 68Q15
ACM classes: F.2; J.4
Cite as: arXiv:2006.15747 [cs.GT]
  (or arXiv:2006.15747v2 [cs.GT] for this version)

Submission history

From: Haris Aziz [view email]
[v1] Sun, 28 Jun 2020 23:48:31 GMT (28kb)
[v2] Tue, 2 Mar 2021 23:28:34 GMT (37kb)

Link back to: arXiv, form interface, contact.