A non-asymptotic analysis of adaptive TD(λ) learning in wireless sensor networks
Wireless sensor network has been widely used in different fields, such as structural health monitoring and artificial intelligence technology. The routing planning, an important part of wireless sensor network, can be formalized as an optimization problem needing to be solved. In this article, a rei...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Hindawi - SAGE Publishing
2022-07-01
|
| Series: | International Journal of Distributed Sensor Networks |
| Online Access: | https://doi.org/10.1177/15501329221114546 |
| _version_ | 1797725515992268800 |
|---|---|
| author | Bing Li Tao Li Muhua Liu Junlong Zhu Mingchuan Zhang Qingtao Wu |
| author_facet | Bing Li Tao Li Muhua Liu Junlong Zhu Mingchuan Zhang Qingtao Wu |
| author_sort | Bing Li |
| collection | DOAJ |
| description | Wireless sensor network has been widely used in different fields, such as structural health monitoring and artificial intelligence technology. The routing planning, an important part of wireless sensor network, can be formalized as an optimization problem needing to be solved. In this article, a reinforcement learning algorithm is proposed to solve the problem of optimal routing in wireless sensor networks, namely, adaptive TD( λ ) learning algorithm referred to as ADTD( λ ) under Markovian noise, which is more practical than i.i.d. (identically and independently distributed) noise in reinforcement learning. Moreover, we also present non-asymptotic analysis of ADTD( λ ) with both constant and diminishing step-sizes. Specifically, when the step-size is constant, the convergence rate of O ( 1 / T ) is achieved, where T is the number of iterations; when the step-size is diminishing, the convergence rate of O ~ ( 1 / T ) is also obtained. In addition, the performance of the algorithm is verified by simulation. |
| first_indexed | 2024-03-12T10:32:23Z |
| format | Article |
| id | doaj.art-172f7f463ad6429faf1ac26d3d5bd8ce |
| institution | Directory Open Access Journal |
| issn | 1550-1477 |
| language | English |
| last_indexed | 2024-03-12T10:32:23Z |
| publishDate | 2022-07-01 |
| publisher | Hindawi - SAGE Publishing |
| record_format | Article |
| series | International Journal of Distributed Sensor Networks |
| spelling | doaj.art-172f7f463ad6429faf1ac26d3d5bd8ce2023-09-02T09:08:21ZengHindawi - SAGE PublishingInternational Journal of Distributed Sensor Networks1550-14772022-07-011810.1177/15501329221114546A non-asymptotic analysis of adaptive TD(λ) learning in wireless sensor networksBing Li0Tao Li1Muhua Liu2Junlong Zhu3Mingchuan Zhang4Qingtao Wu5School of Information Engineering, Henan University of Science and Technology, Luoyang, ChinaDepartment of Information Technology Management, CITIC Heavy Industries Co., Ltd, Luoyang, ChinaSchool of Information Engineering, Henan University of Science and Technology, Luoyang, ChinaSchool of Information Engineering, Henan University of Science and Technology, Luoyang, ChinaSchool of Information Engineering, Henan University of Science and Technology, Luoyang, ChinaSchool of Information Engineering, Henan University of Science and Technology, Luoyang, ChinaWireless sensor network has been widely used in different fields, such as structural health monitoring and artificial intelligence technology. The routing planning, an important part of wireless sensor network, can be formalized as an optimization problem needing to be solved. In this article, a reinforcement learning algorithm is proposed to solve the problem of optimal routing in wireless sensor networks, namely, adaptive TD( λ ) learning algorithm referred to as ADTD( λ ) under Markovian noise, which is more practical than i.i.d. (identically and independently distributed) noise in reinforcement learning. Moreover, we also present non-asymptotic analysis of ADTD( λ ) with both constant and diminishing step-sizes. Specifically, when the step-size is constant, the convergence rate of O ( 1 / T ) is achieved, where T is the number of iterations; when the step-size is diminishing, the convergence rate of O ~ ( 1 / T ) is also obtained. In addition, the performance of the algorithm is verified by simulation.https://doi.org/10.1177/15501329221114546 |
| spellingShingle | Bing Li Tao Li Muhua Liu Junlong Zhu Mingchuan Zhang Qingtao Wu A non-asymptotic analysis of adaptive TD(λ) learning in wireless sensor networks International Journal of Distributed Sensor Networks |
| title | A non-asymptotic analysis of adaptive TD(λ) learning in wireless sensor networks |
| title_full | A non-asymptotic analysis of adaptive TD(λ) learning in wireless sensor networks |
| title_fullStr | A non-asymptotic analysis of adaptive TD(λ) learning in wireless sensor networks |
| title_full_unstemmed | A non-asymptotic analysis of adaptive TD(λ) learning in wireless sensor networks |
| title_short | A non-asymptotic analysis of adaptive TD(λ) learning in wireless sensor networks |
| title_sort | non asymptotic analysis of adaptive td λ learning in wireless sensor networks |
| url | https://doi.org/10.1177/15501329221114546 |
| work_keys_str_mv | AT bingli anonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks AT taoli anonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks AT muhualiu anonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks AT junlongzhu anonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks AT mingchuanzhang anonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks AT qingtaowu anonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks AT bingli nonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks AT taoli nonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks AT muhualiu nonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks AT junlongzhu nonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks AT mingchuanzhang nonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks AT qingtaowu nonasymptoticanalysisofadaptivetdllearninginwirelesssensornetworks |