2022Operations ResearchOpen access

Efficient Fair Division with Minimal Sharing

Fedor Sandomirskiy, Erel Segal-Halevi

Open full text 9 citations

Abstract

When assets are to be divided among several partners, for example, a partnership split, fair division theory can be used to determine a fair allocation. The applicability of existing approaches is limited as they either treat assets as divisible resources that end up being shared among participants or deal with indivisible objects providing only approximate fairness. In practice, sharing is often possible but undesirable, and approximate fairness is not adequate, particularly for highly valuable assets. In “Efficient Fair Division with Minimal Sharing,” Sandomirskiy and Segal-Halevi introduce a novel approach offering a middle ground: the number of shared objects is minimized while maintaining exact fairness and economic efficiency. This minimization can be conducted in polynomial time for generic instances if the number of agents or objects is fixed. Experiments on real data demonstrate a substantial improvement over current methods.

Open-access reader

About this research paper

What this paper is about

When assets are to be divided among several partners, for example, a partnership split, fair division theory can be used to determine a fair allocation. The applicability of existing approaches is limited as they either treat assets as divisible resources that end up being shared among participants or deal with indivisible objects providing only approximate fairness. In practice, sharing is often possible but undesirable, and approximate fairness is not adequate, particularly for highly valuable assets. In “Efficient Fair Division with Minimal Sharing,” Sandomirskiy and Segal-Halevi introduce a novel approach offering a middle ground: the number of shared objects is minimized while maintaining exact fairness and economic efficiency. This minimization can be conducted in polynomial time for generic instances if the number of agents or objects is fixed. Experiments on real data demonstrate a substantial improvement over current methods.

Why it matters

OpenAlex reports 9 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

When assets are to be divided among several partners, for example, a partnership split, fair division theory can be used to determine a fair allocation. The applicability of existing approaches is limited as they either treat assets as divisible resources that end up being shared among participants or deal with indivisible objects providing only approximate fairness. In practice, sharing is often possible but undesirable, and approximate fairness is not adequate, particularly for highly valuable assets. In “Efficient Fair Division with Minimal Sharing,” Sandomirskiy and Segal-Halevi introduce a novel approach offering a middle ground: the number of shared objects is minimized while maintaining exact fairness and economic efficiency. This minimization can be conducted in polynomial time for generic instances if the number of agents or objects is fixed. Experiments on real data demonstrate a substantial improvement over current methods.

Key concepts: Fair division, Division (mathematics), Mathematics, Degenerate energy levels, Pareto principle, Pareto efficiency, Measure (data warehouse), Order (exchange)

Related papers

Back to paper searchBrowse research topicsOriginal source
Efficient Fair Division with Minimal Sharing — Research Paper | ScholarLens