Research map / Federated learning and privacy / Communication-efficient FL

Beyond Flat Gossip: Tiered Gossip Learning for Scalable Collaborative AI

NeurIPS 2026 · Federated learning and privacy › Communication-efficient FL · approach: Decentralized peer-to-peer training

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes Tiered Gossip Learning (TGL), a two-layer push-gossip-pull protocol for decentralized collaborative learning that separates data-holding leaf nodes from a gossip-connected tier of aggregators to scale to thousands of edge devices.

Problem

Centralized federated learning creates server bottlenecks and single points of failure, while flat peer-to-peer gossip learning forces each node's communication degree to grow with network size.

Approach

A two-layer architecture that decouples data-holding leaf nodes from a gossip-connected upper tier, using a push-gossip-pull protocol so per-node communication degree does not need to grow as the network scales.

Experimental setup

Baselines: centralized federated learning flat peer-to-peer gossip learning

Builds on: gossip learning federated learning

Closest papers by meaning

• RelaySum for Decentralized Deep Learning on Heterogeneous Data (NeurIPS 2021 · similarity 0.88)
• Unified Breakdown Analysis for Byzantine Robust Gossip (ICML 2025 · similarity 0.85)
• B-ary Tree Push-Pull Method is Provably Efficient for Distributed Learning on Heterogeneous Data (NeurIPS 2024 · similarity 0.85)
• Unveiling the Power of Multiple Gossip Steps: A Stability-Based Generalization Analysis in Decentralized Training (NeurIPS 2025 · similarity 0.85)
• Robust Distributed Estimation: Extending Gossip Algorithms to Ranking and Trimmed Means (NeurIPS 2025 · similarity 0.84)
• DICE: Data Influence Cascade in Decentralized Learning (ICLR 2025 · similarity 0.84)