Proportionally Fair Matching via Randomized Rounding

Sharmila Duppala,Nathaniel Grammel,Juan Luque,Calum MacRury,Aravind Srinivasan

2024-12-16

Abstract:Given an edge-colored graph, the goal of the proportional fair matching problem is to find a maximum weight matching while ensuring proportional representation (with respect to the number of edges) of each color. The colors may correspond to demographic groups or other protected traits where we seek to ensure roughly equal representation from each group. It is known that, assuming ETH, it is impossible to approximate the problem with $\ell$ colors in time $2^{o(\ell)} n^{\mathcal{O}(1)}$ (i.e., subexponential in $\ell$) even on \emph{unweighted path graphs}. Further, even determining the existence of a non-empty matching satisfying proportionality is NP-Hard. To overcome this hardness, we relax the stringent proportional fairness constraints to a probabilistic notion. We introduce a notion we call $\delta$-\textsc{ProbablyAlmostFair}, where we ensure proportionality up to a factor of at most $(1 \pm \delta)$ for some small $\delta >0$ with high probability. The violation $\delta$ can be brought arbitrarily close to $0$ for some \emph{good} instances with large values of matching size. We propose and analyze simple and fast algorithms for bipartite graphs that achieve constant-factor approximation guarantees, and return a $\delta$-\textsc{ProbablyAlmostFair} matching.

Data Structures and Algorithms

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to ensure proportional fairness in the graph - matching problem, especially when finding the maximum - weight matching in edge - colored graphs, while ensuring that the edges of each color (representing different groups or characteristics) have approximately equal proportions in the matching. However, this problem is extremely computationally difficult. It is even hard to approximate on unweighted path graphs, and determining whether there exists a non - empty matching that satisfies proportional fairness is NP - hard.

To solve this difficult problem, the author introduced a concept of relaxed fairness in a probabilistic sense, called δ - ProbablyFair, that is, with high probability, ensuring that the proportion of edges of each color in the matching does not exceed (1 ± δ) times the target proportion. Through this method, the author designed an effective randomized algorithm for bipartite graphs. While ensuring that the matching weight reaches half of the optimal solution, it can achieve a nearly fair matching result with high probability.

Specifically, this paper mainly solves the following problems:

1. **Define and analyze δ - ProbablyFair matching**: A new fairness metric is proposed, which allows violating strict fairness constraints within a certain range, but can ensure a nearly fair result with high probability.
2. **Design an efficient algorithm**: A simple and fast algorithm for bipartite graphs is proposed, which can find a constant - factor approximate solution in polynomial time and ensure that the matching result is nearly fair in most cases.
3. **Handle special cases**: For the case where there are fairness constraints on only one side (such as α = 0, β < 1), an improved algorithm is proposed, which can obtain a solution close to half of the optimal solution without violating the fairness constraints.

Through these methods, the author not only overcomes the computational complexity of the original problem but also provides a practical solution applicable to multiple practical application scenarios such as advertisement allocation, job recruitment, and crowdsourcing tasks, ensuring that different groups can be treated fairly in these scenarios.