Research & Papers

New algorithm solves election fairness in polynomial time

Computer scientist cracks 26-year-old voting theory problem with O(kmn) operations

Deep Dive

Computer scientist Yizhou Ai presents a new deterministic algorithm for approval-based multiwinner elections that achieves Full Justified Representation (FJR) under both Hare and Droop quota systems in polynomial time. The descending-budget method has voters spread their remaining budgets across representation gaps, with candidates purchased once offers cover a common price. Using a Hare quota price returns a Hare-FJR committee, while a Droop quota price satisfies the more demanding Droop-FJR axiom. Both versions run in O(kmn) rational operations once voter and candidate orders are fixed, sharing a historical-payment invariant and a terminal row–column argument — with the Droop case adding a residual-budget step when all seats are filled.

Key Points
  • Algorithm solves 26-year-old voting theory problem with O(kmn) computational complexity
  • Achieves Full Justified Representation under both Hare (λ_H = n/k) and Droop (λ_D = n/(k+1)) quotas
  • Deterministic solution with historical-payment invariant for both quota systems

Why It Matters

Enables fairer AI-driven decision systems and automated voting protocols in large-scale applications

📬 Get the top 10 AI stories daily