Research Experience

We introduced the Method of Equal Shares with Bounded Overspending (BOS) for participatory budgeting, enhancing the original method by balancing proportionality and efficiency. BOS corrects inefficiencies, particularly in small instances, runs in polynomial time, satisfies key fairness properties, and reliably ensures fair budget allocation.

Working on the systematic analysis of project merging behavior in participatory budgeting elections, focusing on its implications under cooperative game theory. Specifically, the research examines the stability of elections in the absence of merging under different voting rules (e.g., AV/Cost vs. GreedyAV), identifying conditions that deter merging. It also explores constraints on profitable merging through compatibility graphs and their computational complexity (e.g., polynomial for paths but NP-hard for binary trees), as well as voter behavior regarding merged projects. Lastly, the project assesses the social welfare impact of merging, both theoretically and empirically, particularly under the BasicAV rule.

Inspired by the paper of Anari et al. (2023), we focused on optimizing the upper bound of distortion in metric matching with ordinal preferences, resulting in a significant improvement over the previous bound of n² in metric space. We achieved a constant upper bound on distortion for the restricted case of line metric, and an O(n) upper bound for the tree metric.