☆ 4.6 Article

Cross-Silo Federated Learning for Multi-Tier Networks with Vertical and Horizontal Data Partitioning

ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY (2022)

Related references

Note: Only part of the references are listed.

Proceedings Paper Acoustics

MULTI-TIER FEDERATED LEARNING FOR VERTICALLY PARTITIONED DATA

Anirban Das et al.

Summary: This study explores decentralized model training in tiered communication networks with a proposed efficient algorithm suitable for two-tiered networks. The algorithm involves clients performing multiple local gradient steps before sharing updates with hubs, which then adjust coordinates by averaging worker updates and exchange intermediate updates with each other. The approach is further validated through simulation-based experiments with various datasets and objectives.

2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021) (2021)

Add to Collection

Proceedings Paper Computer Science, Information Systems

An Efficient Approach for Cross-Silo Federated Learning to Rank

Yansheng Wang et al.

Summary: Traditional learning-to-rank models are typically trained in a centralized manner, but issues related to data privacy make it difficult to collect data from multiple owners, compromising the performance of the learned models. This paper introduces the CS-F-LTR framework to address efficiency concerns while protecting privacy, and experiments confirm the effectiveness and efficiency of the proposed approach.

2021 IEEE 37TH INTERNATIONAL CONFERENCE ON DATA ENGINEERING (ICDE 2021) (2021)

Add to Collection

Article Computer Science, Information Systems