Decentralized Ranking Aggregation: Gossip Algorithms for Borda and Copeland Consensus

Researchers have developed gossip algorithms that enable decentralized networks to reach consensus on rankings using Borda and Copeland methods without central coordination. The approach allows autonomous agents to compute global ranking consensus through local interactions, with applications in peer-to-peer networks, IoT, and multi-agent systems.

• →New gossip algorithms enable decentralized ranking consensus using classical Borda and Copeland methods without central authority.
• →The approach provides rigorous convergence guarantees and explicit rate bounds for decentralized consensus calculations.
• →Implementation extends to median rank rule and local Kemenization beyond the primary Borda and Copeland methods.
• →Empirical testing shows quick and reliable convergence across various network topologies and ranking datasets.
• →The methodology addresses robustness to corrupted nodes and scalability through reduced communication costs.