EFX Orientations of Multigraphs

Kevin Hsu
2024-10-16
Abstract:We study the fair division of multigraphs with self-loops. In this setting, vertices represent agents and edges represent goods, and a good provides positive utility to an agent only if it is incident with the agent. Whereas previous research has so far only considered simple graphs, we consider the general setting of multigraphs, specifically focusing on the case in which each edge has equal utility to both incident agents, and edges have one of two possible utilities $\alpha > \beta \geq 0$. In contrast with the case of simple graphs for which bipartiteness implies the existence of an EFX orientation, we show that deciding whether a symmetric multigraph $G$ of multiplicity $q \geq 2$ admits an EFX orientation is NP-complete even if $G$ is bipartite, $\alpha > q\beta$, and $G$ contains a structure called a non-trivial odd multitree. Moreover, we show that non-trivial odd multitrees are a forbidden structure in the sense that even very simple non-trivial odd multitrees can fail to admit EFX orientations, and multigraphs that do not contain non-trivial odd multitrees always admit EFX orientations.
Computer Science and Game Theory
What problem does this paper attempt to address?
### What problem does this paper attempt to solve? This paper mainly studies the problem of fair allocation in multigraphs, especially for multigraphs with self - loops. Specifically, vertices represent agents and edges represent goods, and a good only provides positive utility at its associated agent. Different from previous studies that only consider simple graphs, this paper extends to a more general multigraph setting and pays special attention to the situation where each edge has the same utility for the two related agents, and the edge has two possible utility values \( \alpha > \beta \geq 0 \). #### Main research questions: 1. **Existence of EFX allocation**: The author explores whether there is an EFX allocation (Envy - Free up to any good allocation) in multigraphs, that is, allowing an agent \( i \) to envy the allocation of another agent \( j \), but as long as any one good of \( j \) is ignored, \( i \) will no longer envy. 2. **NP - completeness problem**: The author proves that for a symmetric multigraph \( G \) with multiplicity \( q \geq 2 \), even if \( G \) is bipartite, \( \alpha > q\beta \), and \( G \) contains a structure called non - trivial odd multitree, determining whether \( G \) has an EFX allocation is still an NP - complete problem. 3. **Structural influence**: The author points out that the non - trivial odd multitree is the key structure that causes the inability to find an EFX allocation. When a multigraph does not contain this structure, an EFX allocation can always be found, and this allocation can be constructed in polynomial time. #### Specific contributions: - **Hardness result**: It is proved that for a symmetric multigraph with multiplicity \( q \geq 2 \), determining whether it has an EFX allocation is an NP - complete problem. - **Structural analysis**: The concept of non - trivial odd multitree is introduced, and its importance in finding EFX allocations is shown. - **Algorithm design**: A polynomial - time algorithm is proposed to find EFX allocations in multigraphs that do not contain non - trivial odd multitrees. Through these studies, the author not only expands the research scope of the fair allocation problem, but also provides new directions and tools for future research.