Near-Optimal Dropout-Robust Sortition

2025-11-20

AI Summary

Citizen assemblies frequently suffer from unpredictable member attrition, leading to reduced size and diminished demographic representativeness. This paper formalizes the problem as a minimax game and proposes a robust membership selection framework: it pre-optimizes the initial composition of participants—without prior knowledge of the attrition distribution—to ensure both final group size adequacy and demographic representativeness. Methodologically, the approach innovatively integrates projected gradient descent with estimation of the optimal-response attrition distribution, enforcing explicit fairness constraints on each candidate’s inclusion probability. Experiments on real-world datasets demonstrate that the algorithm achieves an approximate Pareto-optimal trade-off among representational bias, robustness to attrition, and selection fairness, significantly outperforming existing baseline methods.

Technology Category

• Knowledge Representation and Reasoning: Preferences
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Citizens' assemblies - small panels of citizens that convene to deliberate on policy issues - often face the issue of panelists dropping out at the last-minute. Without intervention, these dropouts compromise the size and representativeness of the panel, prompting the question: Without seeing the dropouts ahead of time, can we choose panelists such that after dropouts, the panel will be representative and appropriately-sized? We model this problem as a minimax game: the minimizer aims to choose a panel that minimizes the loss, i.e., the deviation of the ultimate panel from predefined representation targets. Then, an adversary defines a distribution over dropouts from which the realized dropouts are drawn. Our main contribution is an efficient loss-minimizing algorithm, which remains optimal as we vary the maximizer's power from worst case to average case. Our algorithm - which iteratively plays a projected gradient descent subroutine against an efficient algorithm for computing the best-response dropout distribution - also addresses a key open question in the area: how to manage dropouts while ensuring that each potential panelist is chosen with relatively equal probabilities. Using real-world datasets, we compare our algorithms to existing benchmarks, and we offer the first characterizations of tradeoffs between robustness, loss, and equality in this problem.

Problem

• Addressing last-minute dropouts in citizens' assemblies to maintain panel representativeness
• Developing algorithms to minimize deviation from representation targets after dropouts
• Balancing robustness, loss, and equality in panel selection under dropout uncertainty

Innovation

• Efficient loss-minimizing algorithm for dropout-robust sortition
• Projected gradient descent against best-response dropout distribution
• Balancing robustness, loss, and equality in panel selection