Decentralized non-convex online optimization with adaptive momentum estimation and quantized communication
Abstract In this work, we consider the decentralized non-convex online optimization problem over an undirected network. To solve the problem over a communication-efficient manner, we propose a novel quantized decentralized adaptive momentum gradient descent algorithm based on the adaptive momentum e...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-03-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-025-01818-8 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850065316818714624 |
|---|---|
| author | Yunshan Lv Hailing Xiong Fuqing Zhang Shengying Dong Xiangguang Dai |
| author_facet | Yunshan Lv Hailing Xiong Fuqing Zhang Shengying Dong Xiangguang Dai |
| author_sort | Yunshan Lv |
| collection | DOAJ |
| description | Abstract In this work, we consider the decentralized non-convex online optimization problem over an undirected network. To solve the problem over a communication-efficient manner, we propose a novel quantized decentralized adaptive momentum gradient descent algorithm based on the adaptive momentum estimation methods, where quantified information is exchanged between agents. The proposed algorithm not only can effectively reduce the data transmission volume but also contribute to improved convergence. Theoretical analysis proves that the proposed algorithm can achieve sublinear dynamic regret under appropriate step-size and quantization level, which matches the convergence of the decentralized online algorithm with exact-communication. Extensive simulations are given to demonstrate the efficacy of the algorithm. |
| format | Article |
| id | doaj-art-bfa604ab1e3e444bb9cc505e7ad3f099 |
| institution | DOAJ |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-bfa604ab1e3e444bb9cc505e7ad3f0992025-08-20T02:49:02ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-03-0111411810.1007/s40747-025-01818-8Decentralized non-convex online optimization with adaptive momentum estimation and quantized communicationYunshan Lv0Hailing Xiong1Fuqing Zhang2Shengying Dong3Xiangguang Dai4College of Computer and Information Science, Southwest UniversityCollege of Electronic and Information Engineering, Southwest UniversityAerospace Information Research Institute, Chinese Academy of SciencesCollege of Big Data, Chongqing College of Mobile CommunicationSchool of Three Gorges Artificial Intelligence, Chongqing Three Gorges UniversityAbstract In this work, we consider the decentralized non-convex online optimization problem over an undirected network. To solve the problem over a communication-efficient manner, we propose a novel quantized decentralized adaptive momentum gradient descent algorithm based on the adaptive momentum estimation methods, where quantified information is exchanged between agents. The proposed algorithm not only can effectively reduce the data transmission volume but also contribute to improved convergence. Theoretical analysis proves that the proposed algorithm can achieve sublinear dynamic regret under appropriate step-size and quantization level, which matches the convergence of the decentralized online algorithm with exact-communication. Extensive simulations are given to demonstrate the efficacy of the algorithm.https://doi.org/10.1007/s40747-025-01818-8Non-convex online optimizationDecentralized algorithmQuantized communicationAdaptive momentum estimationSublinear regret |
| spellingShingle | Yunshan Lv Hailing Xiong Fuqing Zhang Shengying Dong Xiangguang Dai Decentralized non-convex online optimization with adaptive momentum estimation and quantized communication Complex & Intelligent Systems Non-convex online optimization Decentralized algorithm Quantized communication Adaptive momentum estimation Sublinear regret |
| title | Decentralized non-convex online optimization with adaptive momentum estimation and quantized communication |
| title_full | Decentralized non-convex online optimization with adaptive momentum estimation and quantized communication |
| title_fullStr | Decentralized non-convex online optimization with adaptive momentum estimation and quantized communication |
| title_full_unstemmed | Decentralized non-convex online optimization with adaptive momentum estimation and quantized communication |
| title_short | Decentralized non-convex online optimization with adaptive momentum estimation and quantized communication |
| title_sort | decentralized non convex online optimization with adaptive momentum estimation and quantized communication |
| topic | Non-convex online optimization Decentralized algorithm Quantized communication Adaptive momentum estimation Sublinear regret |
| url | https://doi.org/10.1007/s40747-025-01818-8 |
| work_keys_str_mv | AT yunshanlv decentralizednonconvexonlineoptimizationwithadaptivemomentumestimationandquantizedcommunication AT hailingxiong decentralizednonconvexonlineoptimizationwithadaptivemomentumestimationandquantizedcommunication AT fuqingzhang decentralizednonconvexonlineoptimizationwithadaptivemomentumestimationandquantizedcommunication AT shengyingdong decentralizednonconvexonlineoptimizationwithadaptivemomentumestimationandquantizedcommunication AT xiangguangdai decentralizednonconvexonlineoptimizationwithadaptivemomentumestimationandquantizedcommunication |