Three Candidate Plurality is Stablest for Small Correlations
Steven Heilman, Alex Tarter

TL;DR
This paper proves a structure theorem for noise stable partitions in Gaussian space, confirming the Plurality is Stablest Conjecture for three candidates under certain conditions and providing new variational proofs for related inequalities.
Contribution
It establishes a dimension reduction result for noise stable partitions and proves the Plurality is Stablest Conjecture for three candidates with fixed correlation, advancing understanding of optimal voting schemes.
Findings
Proves the Plurality is Stablest Conjecture for three candidates for all small correlations.
Provides a variational proof of Borell's Inequality.
Establishes a structure theorem for noise stable partitions in Gaussian space.
Abstract
Using the calculus of variations, we prove the following structure theorem for noise stable partitions: a partition of -dimensional Euclidean space into disjoint sets of fixed Gaussian volumes that maximize their noise stability must be -dimensional, if . In particular, the maximum noise stability of a partition of sets in of fixed Gaussian volumes is constant for all satisfying . From this result, we obtain: (i) A proof of the Plurality is Stablest Conjecture for candidate elections, for all correlation parameters satisfying , where is a fixed constant (that does not depend on the dimension ), when each candidate has an equal chance of winning. (ii) A variational proof of Borell's Inequality (corresponding to the case ). The structure theorem answers a question of…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
