Coordinate-Descent Adaptation over Hamiltonian Multi-Agent Networks
The incremental least-mean-square (ILMS) algorithm is a useful method to perform distributed adaptation and learning in Hamiltonian networks. To implement the ILMS algorithm, each node needs to receive the local estimate of the previous node on the cycle path to update its own local estimate. Howeve...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/22/7732 |
| _version_ | 1797508468792360960 |
|---|---|
| author | Azam Khalili Vahid Vahidpour Amir Rastegarnia Ali Farzamnia Kenneth Teo Tze Kin Saeid Sanei |
| author_facet | Azam Khalili Vahid Vahidpour Amir Rastegarnia Ali Farzamnia Kenneth Teo Tze Kin Saeid Sanei |
| author_sort | Azam Khalili |
| collection | DOAJ |
| description | The incremental least-mean-square (ILMS) algorithm is a useful method to perform distributed adaptation and learning in Hamiltonian networks. To implement the ILMS algorithm, each node needs to receive the local estimate of the previous node on the cycle path to update its own local estimate. However, in some practical situations, perfect data exchange may not be possible among the nodes. In this paper, we develop a new version of ILMS algorithm, wherein in its adaptation step, only a random subset of the coordinates of update vector is available. We draw a comparison between the proposed coordinate-descent incremental LMS (CD-ILMS) algorithm and the ILMS algorithm in terms of convergence rate and computational complexity. Employing the energy conservation relation approach, we derive closed-form expressions to describe the learning curves in terms of excess mean-square-error (EMSE) and mean-square deviation (MSD). We show that, the CD-ILMS algorithm has the same steady-state error performance compared with the ILMS algorithm. However, the CD-ILMS algorithm has a faster convergence rate. Numerical examples are given to verify the efficiency of the CD-ILMS algorithm and the accuracy of theoretical analysis. |
| first_indexed | 2024-03-10T05:04:25Z |
| format | Article |
| id | doaj.art-bf739746f80e4649b03e778d174876a0 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T05:04:25Z |
| publishDate | 2021-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-bf739746f80e4649b03e778d174876a02023-11-23T01:28:52ZengMDPI AGSensors1424-82202021-11-012122773210.3390/s21227732Coordinate-Descent Adaptation over Hamiltonian Multi-Agent NetworksAzam Khalili0Vahid Vahidpour1Amir Rastegarnia2Ali Farzamnia3Kenneth Teo Tze Kin4Saeid Sanei5Department of Electrical Engineering, Malayer University, Malayer 65719-95863, IranDepartment of Electrical Engineering, Malayer University, Malayer 65719-95863, IranDepartment of Electrical Engineering, Malayer University, Malayer 65719-95863, IranFaculty of Engineering, Universiti Malaysia Sabah, Kota Kinabalu 88400, MalaysiaFaculty of Engineering, Universiti Malaysia Sabah, Kota Kinabalu 88400, MalaysiaScience and Technology, Nottingham Trent University, Clifton Lane, Nottingham NG11 8NS, UKThe incremental least-mean-square (ILMS) algorithm is a useful method to perform distributed adaptation and learning in Hamiltonian networks. To implement the ILMS algorithm, each node needs to receive the local estimate of the previous node on the cycle path to update its own local estimate. However, in some practical situations, perfect data exchange may not be possible among the nodes. In this paper, we develop a new version of ILMS algorithm, wherein in its adaptation step, only a random subset of the coordinates of update vector is available. We draw a comparison between the proposed coordinate-descent incremental LMS (CD-ILMS) algorithm and the ILMS algorithm in terms of convergence rate and computational complexity. Employing the energy conservation relation approach, we derive closed-form expressions to describe the learning curves in terms of excess mean-square-error (EMSE) and mean-square deviation (MSD). We show that, the CD-ILMS algorithm has the same steady-state error performance compared with the ILMS algorithm. However, the CD-ILMS algorithm has a faster convergence rate. Numerical examples are given to verify the efficiency of the CD-ILMS algorithm and the accuracy of theoretical analysis.https://www.mdpi.com/1424-8220/21/22/7732adaptive estimationcoordinate-descentdistributed networksincremental algorithm |
| spellingShingle | Azam Khalili Vahid Vahidpour Amir Rastegarnia Ali Farzamnia Kenneth Teo Tze Kin Saeid Sanei Coordinate-Descent Adaptation over Hamiltonian Multi-Agent Networks Sensors adaptive estimation coordinate-descent distributed networks incremental algorithm |
| title | Coordinate-Descent Adaptation over Hamiltonian Multi-Agent Networks |
| title_full | Coordinate-Descent Adaptation over Hamiltonian Multi-Agent Networks |
| title_fullStr | Coordinate-Descent Adaptation over Hamiltonian Multi-Agent Networks |
| title_full_unstemmed | Coordinate-Descent Adaptation over Hamiltonian Multi-Agent Networks |
| title_short | Coordinate-Descent Adaptation over Hamiltonian Multi-Agent Networks |
| title_sort | coordinate descent adaptation over hamiltonian multi agent networks |
| topic | adaptive estimation coordinate-descent distributed networks incremental algorithm |
| url | https://www.mdpi.com/1424-8220/21/22/7732 |
| work_keys_str_mv | AT azamkhalili coordinatedescentadaptationoverhamiltonianmultiagentnetworks AT vahidvahidpour coordinatedescentadaptationoverhamiltonianmultiagentnetworks AT amirrastegarnia coordinatedescentadaptationoverhamiltonianmultiagentnetworks AT alifarzamnia coordinatedescentadaptationoverhamiltonianmultiagentnetworks AT kennethteotzekin coordinatedescentadaptationoverhamiltonianmultiagentnetworks AT saeidsanei coordinatedescentadaptationoverhamiltonianmultiagentnetworks |