View in EDS

Attention-Based Adaptive Reinforcement Learning for Efficient Resource Management in Edge Computing.

Saved in:
Bibliographic Details
Title: Attention-Based Adaptive Reinforcement Learning for Efficient Resource Management in Edge Computing.
Authors: Hameed, Shabi Alam1 (AUTHOR) shabiazam@gmail.com, Farid, Muhammad Shahid Ghulam1 (AUTHOR)
Source: Cybernetics & Systems. 2026, Vol. 57 Issue 4, p603-640. 38p.
Subjects: Resource management, Edge computing, Reinforcement learning, Optimization algorithms, Deep learning, Mathematical optimization, Real-time computing
Abstract: The proposed system uses an advanced deep learning method called Adaptive Deep Deterministic Policy Gradient with Attention (Ada-DDPG-A) to intelligently allocate tasks and resources. Later, by using the Fitness Oriented African Bison Optimization Algorithm (FE-ABOA), the developed deep learning model Ada-DDPG-A is precisely tuned, leading to much better resource management. An adaptive component in this novel model learns and adjusts its behavior in real-time as it receives new data. This hybrid approach optimizes performance, improves resource utilization, and enhances overall system reliability. This approach effectively solves the NP-hard resource management problem and enhances task effectiveness through precise parameter tuning. Finally, several validations are executed in the suggested technique to observe its efficiency over the existing techniques. In time slot 1, the delay time of the developed FE-ABOA-Ada-DDPG-A model is 66.43 ms, whereas the delay times for OOA-Ada-DDPG-A, MAO-Ada-DDPG-A, SCO-Ada-DDPG-A, and ABOA-Ada-DDPG-A are 74.74, 147.97, 99.27, and 137.57 ms, respectively. These results demonstrate that the developed approach significantly minimizes latency, outperforming existing algorithms in real-time processing efficiency. It conclusively shows that the enhancements made to the algorithm result in a substantial and meaningful improvement over established optimization techniques. [ABSTRACT FROM AUTHOR]
Copyright of Cybernetics & Systems is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:The proposed system uses an advanced deep learning method called Adaptive Deep Deterministic Policy Gradient with Attention (Ada-DDPG-A) to intelligently allocate tasks and resources. Later, by using the Fitness Oriented African Bison Optimization Algorithm (FE-ABOA), the developed deep learning model Ada-DDPG-A is precisely tuned, leading to much better resource management. An adaptive component in this novel model learns and adjusts its behavior in real-time as it receives new data. This hybrid approach optimizes performance, improves resource utilization, and enhances overall system reliability. This approach effectively solves the NP-hard resource management problem and enhances task effectiveness through precise parameter tuning. Finally, several validations are executed in the suggested technique to observe its efficiency over the existing techniques. In time slot 1, the delay time of the developed FE-ABOA-Ada-DDPG-A model is 66.43 ms, whereas the delay times for OOA-Ada-DDPG-A, MAO-Ada-DDPG-A, SCO-Ada-DDPG-A, and ABOA-Ada-DDPG-A are 74.74, 147.97, 99.27, and 137.57 ms, respectively. These results demonstrate that the developed approach significantly minimizes latency, outperforming existing algorithms in real-time processing efficiency. It conclusively shows that the enhancements made to the algorithm result in a substantial and meaningful improvement over established optimization techniques. [ABSTRACT FROM AUTHOR]
ISSN:01969722
DOI:10.1080/01969722.2026.2643176