# Delta Sum Learning: an approach for fast and global convergence in Gossip Learning

arXiv: https://arxiv.org/abs/2512.01549 · HTML: https://mlsanity.com/paper/2512.01549/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 25 references that are in the map.
- First posted on arXiv 1 Dec 2025.

## Builds on (most cited first, 9 of 9)

- [Federated Learning on Non-IID Data: A Survey](https://mlsanity.com/paper/2106.06843/) (2021)
- [Distributionally Robust Federated Averaging](https://mlsanity.com/paper/2102.12660/) (2021)
- [Decentralized Federated Learning: A Survey and Perspective](https://mlsanity.com/paper/2306.01603/) (2023)
- [Byzantine-Robust Decentralized Federated Learning](https://mlsanity.com/paper/2406.10416/) (2024)
- [Throughput-Optimal Topology Design for Cross-Silo Federated Learning](https://mlsanity.com/paper/2010.12229/) (2020)
- [Fedstellar: A Platform for Decentralized Federated Learning](https://mlsanity.com/paper/2306.09750/) (2023)
- [OpenEI: An Open Framework for Edge Intelligence](https://mlsanity.com/paper/1906.01864/) (2019)
- [Vanishing Variance Problem in Fully Decentralized Neural-Network Systems](https://mlsanity.com/paper/2404.04616/) (2024)
- [Age of Gossip with the Push-Pull Protocol](https://mlsanity.com/paper/2409.20490/) (2024)

## Abstract (arXiv)

Federated Learning is a popular approach for distributed learning due to its security and computational benefits. With the advent of powerful devices in the network edge, Gossip Learning further decentralizes Federated Learning by removing centralized integration and relying fully on peer to peer updates. However, the averaging methods generally used in both Federated and Gossip Learning are not ideal for model accuracy and global convergence. Additionally, there are few options to deploy Learning workloads in the edge as part of a larger application using a declarative approach such as Kubernetes manifests. This paper proposes Delta Sum Learning as a method to improve the basic aggregation operation in Gossip Learning, and implements it in a decentralized orchestration framework based on Open Application Model, which allows for dynamic node discovery and intent-driven deployment of multi-workload applications. Evaluation results show that Delta Sum performance is on par with alternative integration methods for 10 node topologies, but results in a 58% lower global accuracy drop when scaling to 50 nodes. Overall, it shows strong global convergence and a logarithmic loss of accuracy with increasing topology size compared to a linear loss for alternatives under limited connectivity.