Applications with adaptability are widely available on the edge devices with energy harvesting capabilities. For their runtime quality optimization, however, current approaches can not tackle the variations of quality modeling and harvested energy simultaneously. Therefore, in this paper, we are the first to propose a deep reinforcement learning (DRL)-based DVFS method that optimizes the application execution quality of energy harvesting edge devices to mitigate the variations. First, we propose a baseline DRL formulation that novelly migrates the objective of quality maximization into a reward function and constructs a DRL quality agent. Second, we devise a long short-term memory (LSTM)-based selector that performs DRL quality agent selection based on the energy harvesting history. Third, we further propose two optimization methods to alleviate the non-negligible overhead of DRL computations: 1) an improved thinking-while-moving concurrent DRL scheme to compromise the ‘state drifting’ issue during the DRL decision process, and 2) a variable inter-state duration decision scheme that compromises the DVFS overhead incurred in each action taken. The experiments take an adaptive stereo matching application as a case study. The results show that the proposed DRL-based DVFS method on average achieves 17.9% runtime reduction and 22.05% quality improvement compared to state-of-the-art solutions.
|Journal||IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems|
|Publication status||Accepted/In press - 2022|
- Adaptation models
- adaptive application
- Computational modeling
- deep reinforcement learning
- Energy harvesting
- energy harvesting.
- Quality optimization
- stereo matching
- Task analysis
ASJC Scopus subject areas
- Computer Graphics and Computer-Aided Design
- Electrical and Electronic Engineering