Abstract:We study fair division of indivisible mixed manna when agents have unequal entitlements, with weighted envy-freeness up to one item (WEF1) as our primary notion of fairness. We identify several shortcomings of existing techniques to achieve WEF1. Hence, we relax WEF1 to weighted envy-freeness up to 1 transfer (WEF1T), and give a polynomial-time algorithm for achieving it. We also generalize Fisher markets to the mixed manna setting, and use them to get a polynomial-time algorithm for two agents that outputs a WEF1 allocation.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is how to achieve fair distribution in resource allocation when the subjects participating in the allocation (such as people or agents) have different rights (i.e., unequal interests) and the resources are of mixed types (including both goods with positive values and tasks with negative values). Specifically, the paper focuses on achieving the fairness criterion of "Weighted Envy - Free up to One item (WEF1)" under these conditions.
### Specific Problem Description
1. **Fair Allocation of Mixed Resources**:
- The paper studies the fair allocation problem of indivisible mixed resources (Mixed Manna). The mixed resources here refer to both goods with positive values and tasks with negative values. Each participant's value assessment of the same resource may be subjective, that is, for some people, a certain task may be regarded as a good, while for others it may be regarded as a task.
2. **Unequal Rights**:
- The entitlements of participants can be unequal. For example, in the distribution of an estate, different relatives may receive different shares according to their kinship; or in a cooperative project, investors obtain different return ratios according to the amount of investment.
3. **Existing Technical Limitations**:
- The existing techniques for achieving WEF1 have some deficiencies, especially in dealing with mixed resources and unequal entitlements. Therefore, the author introduces a slightly relaxed fairness criterion, called "Weighted Envy - Free up to One Transfer (WEF1T)".
### Solution
- **Algorithm Design**:
- The author proposes a polynomial - time algorithm that can achieve the WEF1T allocation. In addition, they also generalize the concept of Fisher market equilibrium to the mixed - resource setting, thus providing a polynomial - time algorithm for two participants, which can achieve WEF1 while reaching fractional Pareto optimality (fPO).
### Conclusion
- The paper shows that in the case of mixed resources and unequal entitlements, achieving a strict WEF1 allocation is a challenging problem. However, by introducing the slightly relaxed standard of WEF1T, a fair allocation scheme can still be found within polynomial time. For the case of two participants, not only can WEF1 be achieved, but fPO can also be reached simultaneously.
### Open Problems
- Whether there is a method to achieve WEF1 allocation in the case of three or more participants, even under unequal entitlements, is still an unsolved problem.
- For three or more participants, whether there is always an EF1 + PO allocation, even in the case of equal entitlements, is still an unknown area.
This paper provides new solutions to the fair allocation problem under the conditions of mixed resources and unequal entitlements by introducing new fairness criteria and algorithms, and also points out the directions for future research.