# Peer-to-Peer Learning + Consensus with Non-IID Data

arXiv: https://arxiv.org/abs/2312.13602 · HTML: https://mlsanity.com/paper/2312.13602/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 17 references that are in the map.
- First posted on arXiv 21 Dec 2023.

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

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Communication-Efficient Learning of Deep Networks from Decentralized Data](https://mlsanity.com/paper/1602.05629/) (2016)
- [Federated Learning with Non-IID Data](https://mlsanity.com/paper/1806.00582/) (2018)
- [Can Decentralized Algorithms Outperform Centralized Algorithms? A Case Study for Decentralized Parallel Stochastic Gradient Descent](https://mlsanity.com/paper/1705.09056/) (2017)
- [A Unified Theory of Decentralized SGD with Changing Topology and Local Updates](https://mlsanity.com/paper/2003.10422/) (2020)
- [Federated Learning with Cooperating Devices: A Consensus Approach for Massive IoT Networks](https://mlsanity.com/paper/1912.13163/) (2019)
- [Collaborative Deep Learning in Fixed Topology Networks](https://mlsanity.com/paper/1706.07880/) (2017)
- [Enable Deep Learning on Mobile Devices: Methods, Systems, and Applications](https://mlsanity.com/paper/2204.11786/) (2022)
- [Peer-to-Peer Deep Learning for Beyond-5G IoT](https://mlsanity.com/paper/2310.18861/) (2023)

## Built on by (most cited first, 3 of 3)

- [Peer-to-Peer Learning Dynamics of Wide Neural Networks](https://mlsanity.com/paper/2409.15267/) (2024)
- [FedBaF: Federated Learning Aggregation Biased by a Foundation Model](https://mlsanity.com/paper/2410.18352/) (2024)
- [GLUE: Gradient-free Learning to Unify Experts](https://mlsanity.com/paper/2512.22467/) (2025)

## Abstract (arXiv)

Peer-to-peer deep learning algorithms are enabling distributed edge devices to collaboratively train deep neural networks without exchanging raw training data or relying on a central server. Peer-to-Peer Learning (P2PL) and other algorithms based on Distributed Local-Update Stochastic/mini-batch Gradient Descent (local DSGD) rely on interleaving epochs of training with distributed consensus steps. This process leads to model parameter drift/divergence amongst participating devices in both IID and non-IID settings. We observe that model drift results in significant oscillations in test performance evaluated after local training and consensus phases. We then identify factors that amplify performance oscillations and demonstrate that our novel approach, P2PL with Affinity, dampens test performance oscillations in non-IID settings without incurring any additional communication cost.