Abstract:Impartial selection problems are concerned with the selection of one or more agents from a set based on mutual nominations from within the set. To avoid strategic nominations of the agents, the axiom of impartiality requires that the selection of each agent is independent of the nominations cast by that agent. This paper initiates the study of impartial selection problems where the nominations are weighted and the set of agents that can be selected is restricted by a combinatorial constraint. We call a selection mechanism $\alpha$-optimal if, for every instance, the ratio between the total sum of weighted nominations of the selected set and that of the best feasible set of agents is at least $\alpha$. We show that a natural extension of a mechanism studied for the selection of a single agent remains impartial and $\frac{1}{4}$-optimal for general independence systems, and we generalize upper bounds from the selection of multiple agents by parameterizing them by the girth of the independence system. We then focus on independence systems defined by knapsack and matroid constraints, giving impartial mechanisms that exploit a greedy order of the agents and achieve approximation ratios of $\frac{1}{3}$ and $\frac{1}{2}$, respectively, when agents cast a single nomination. For graphic matroids, we further devise an impartial and $\frac{1}{3}$-optimal mechanism for an arbitrary number of unweighted nominations.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the fair selection problem under combinatorial constraints. Specifically, the author has studied how to make a fair selection among a group of agents in a given independence system while considering weighted nominations. To ensure the fairness of the selection process, the selection probability of each agent should not be affected by the scores it gives to other agents. In addition, the set of selected agents needs to satisfy specific combinatorial constraints. ### Background of the Fair Selection Problem The core of the fair selection problem lies in avoiding strategic nominations, that is, an agent should not increase its own chance of being selected by manipulating the scores it gives to other agents. To solve this problem, the author introduced the **axiom of impartiality**, which requires that the selection of each agent is independent of its own nominations. ### Research Objectives The main objective of this paper is to study the performance of fair selection mechanisms in a broader context (such as weighted nominations and combinatorial constraints). Specifically: 1. **Expand Existing Models**: Generalize the existing fair selection models to more general combinatorial constraints, such as knapsack constraints and matroid constraints. 2. **Allow Weighted Nominations**: Not be limited to binary nominations, but also allow agents to submit weighted nominations. 3. **Evaluate Mechanism Performance**: Study the approximation ratio of these extended fair selection mechanisms, that is, the ratio between the total weight of the selected agent set and the total weight of the optimal feasible set. ### Main Contributions 1. **2 - Partition Mechanism for General Independence Systems**: It is proved that the 2 - partition mechanism proposed by Alon et al. still maintains an approximation ratio of $\frac{1}{4}$ in a broader situation. 2. **Upper Bound Based on Girth**: For 1 - sparse instances, an upper bound of $1-\frac{1}{g(g - 1)}$ depending on the girth (the size of the minimum dependent set) of the independence system is given; for any instance, a more general upper bound of $1-\frac{1}{2g + 1}$ is given. 3. **Mechanism Under Knapsack Constraints**: A randomized mechanism is designed, which achieves an approximation ratio of $\frac{1}{d + 2}$ in d - sparse instances. In particular, it achieves an approximation ratio of $\frac{1}{3}$ when each agent evaluates at most one other agent. 4. **Mechanism Under Matroid Constraints**: A randomized mechanism is proposed, which achieves an approximation ratio of $\frac{1}{2}$ in 1 - sparse instances and it is proved that this is the optimal general guarantee. 5. **Mechanism Under Simple Graph Matroids**: For the case of binary scores, a $\frac{1}{3}$ - optimal fair selection mechanism is given. ### Conclusions By introducing new methods and techniques, this paper significantly expands the research scope of the fair selection problem and provides several important results, especially the performance analysis under different types of combinatorial constraints. This provides a theoretical basis for designing more fair and efficient review mechanisms in practical applications in the future.

Impartial Selection Under Combinatorial Constraints

Deterministic Impartial Selection with Weights

Impartial Selection with Additive Guarantees via Iterated Deletion

Improved Bounds for Single-Nomination Impartial Selection

Optimal Impartial Correspondences

Fair integer programming under dichotomous and cardinal preferences

Impartial Rank Aggregation

Fairness and efficiency in strategy-proof object allocation mechanisms

Online Combinatorial Assignment in Independence Systems

Fair, Manipulation-Robust, and Transparent Sortition

Subset Selection Based On Multiple Rankings in the Presence of Bias: Effectiveness of Fairness Constraints for Multiwinner Voting Score Functions

Algorithmically Fair Maximization of Multiple Submodular Objective Functions

On Incentive Compatible Competitive Selection Protocols.

Combinatorial Secretary Problems with Ordinal Information

On a Competitive Secretary Problem with Deferred Selections

Near-Optimal Auctions on Independence Systems

Balancing Notions of Equity: Trade-offs Between Fair Portfolio Sizes and Achievable Guarantees

Long-Term Fairness in Sequential Multi-Agent Selection with Positive Reinforcement

Impartial games with entailing moves

Optimal algorithms for multiwinner elections and the Chamberlin–Courant Rule

Constrained Non-Monotone Submodular Maximization: Offline and Secretary Algorithms