Federated Learning on Non-IID Data: A Survey

By Hangyu Zhu, Jinjin Xu, Shiqing Liu, Yaochu Jin. Neurocomputing, 2021

1,345 citations recorded by the source corpus.

Federated learning is an emerging distributed machine learning framework for privacy preservation. However, models trained in federated learning usually have worse performance than those trained in the standard centralized learning mode, especially when the training data are not independent and identically distributed (Non-IID) on the local devices. In this survey, we pro-vide a detailed analysis of the influence of Non-IID data on both parametric and non-parametric machine learning models in both horizontal and vertical federated learning. In addition, cur-rent research work on handling challenges of Non-IID data in federated learning are reviewed, and both advantages and disadvantages of these approaches are discussed. Finally, we suggest several future research directions before concluding the paper.

Subfields: Artificial Intelligence, Building and Construction, Computer Science Applications

View Paper: https://www.semanticscholar.org/paper/81c3c02f8d82c826833c7ac74746f65564930feb

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Bo Du
2. Mang Ye
3. Pong Chi Yuen
4. Dacheng Tao
5. Mohammad R. Hayajneh
6. Haftay Gebreslasie Abreha
7. Xiuwen Fang
8. Mohamed Adel Serhani

In Building and Construction

1. Dacheng Tao
2. Bo Du
3. Mang Ye

In Computer Science Applications

1. Dacheng Tao
2. Bo Du
3. Mang Ye

Citations in the corpus

• Federated Learning in Edge Computing: A Systematic Survey (2022, result): “In contrast to traditional DL training, participating devices across all edge networks generate and collect datasets in a non-Identical and Independent Distribution (non-IID) in FL model training, which has the potential to bias optimization procedures in DL model development [182].”
• Heterogeneous Federated Learning: State-of-the-art and Research Challenges (2023, background): “1) Statistical heterogeneity: the collected data in different participants may be Non-Independent Identically Distributed (Non-IID) [257] or unbalanced, resulting in inconsistent update optimization directions of participants.”