QSFL: A Two-Level Uplink Communication Optimization Framework for Federated Learning

Liping Yi, Wang Gang, Liu Xiaoguang
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:25501-25513, 2022.

Abstract

In cross-device Federated Learning (FL), the communication cost of transmitting full-precision models between edge devices and a central server is a significant bottleneck, due to expensive, unreliable, and low-bandwidth wireless connections. As a solution, we propose a novel FL framework named QSFL, towards optimizing FL uplink (client-to-server) communication at both client and model levels. At the client level, we design a Qualification Judgment (QJ) algorithm to sample high-qualification clients to upload models. At the model level, we explore a Sparse Cyclic Sliding Segment (SCSS) algorithm to further compress transmitted models. We prove that QSFL can converge over wall-to-wall time, and develop an optimal hyperparameter searching algorithm based on theoretical analysis to enable QSFL to make the best trade-off between model accuracy and communication cost. Experimental results show that QSFL achieves state-of-the-art compression ratios with marginal model accuracy degradation.

Cite this Paper


BibTeX
@InProceedings{pmlr-v162-yi22a, title = {{QSFL}: A Two-Level Uplink Communication Optimization Framework for Federated Learning}, author = {Yi, Liping and Gang, Wang and Xiaoguang, Liu}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {25501--25513}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/yi22a/yi22a.pdf}, url = {https://proceedings.mlr.press/v162/yi22a.html}, abstract = {In cross-device Federated Learning (FL), the communication cost of transmitting full-precision models between edge devices and a central server is a significant bottleneck, due to expensive, unreliable, and low-bandwidth wireless connections. As a solution, we propose a novel FL framework named QSFL, towards optimizing FL uplink (client-to-server) communication at both client and model levels. At the client level, we design a Qualification Judgment (QJ) algorithm to sample high-qualification clients to upload models. At the model level, we explore a Sparse Cyclic Sliding Segment (SCSS) algorithm to further compress transmitted models. We prove that QSFL can converge over wall-to-wall time, and develop an optimal hyperparameter searching algorithm based on theoretical analysis to enable QSFL to make the best trade-off between model accuracy and communication cost. Experimental results show that QSFL achieves state-of-the-art compression ratios with marginal model accuracy degradation.} }
Endnote
%0 Conference Paper %T QSFL: A Two-Level Uplink Communication Optimization Framework for Federated Learning %A Liping Yi %A Wang Gang %A Liu Xiaoguang %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-yi22a %I PMLR %P 25501--25513 %U https://proceedings.mlr.press/v162/yi22a.html %V 162 %X In cross-device Federated Learning (FL), the communication cost of transmitting full-precision models between edge devices and a central server is a significant bottleneck, due to expensive, unreliable, and low-bandwidth wireless connections. As a solution, we propose a novel FL framework named QSFL, towards optimizing FL uplink (client-to-server) communication at both client and model levels. At the client level, we design a Qualification Judgment (QJ) algorithm to sample high-qualification clients to upload models. At the model level, we explore a Sparse Cyclic Sliding Segment (SCSS) algorithm to further compress transmitted models. We prove that QSFL can converge over wall-to-wall time, and develop an optimal hyperparameter searching algorithm based on theoretical analysis to enable QSFL to make the best trade-off between model accuracy and communication cost. Experimental results show that QSFL achieves state-of-the-art compression ratios with marginal model accuracy degradation.
APA
Yi, L., Gang, W. & Xiaoguang, L.. (2022). QSFL: A Two-Level Uplink Communication Optimization Framework for Federated Learning. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:25501-25513 Available from https://proceedings.mlr.press/v162/yi22a.html.

Related Material