Q-DPM: An Efficient Model-Free Dynamic Power Management Technique

Authors:
Min Li;Xiaobo Wu;Richard Yao;Xiaolang Yan
Affiliations:
Institute of VLSI Design, Zhejiang Univ., China/ Microsoft Research Asia;Institute of VLSI Design, Zhejiang Univ., China;Microsoft Research Asia;Institute of VLSI Design, Zhejiang Univ., China
Venue:
Proceedings of the conference on Design, Automation and Test in Europe - Volume 1
Year:
2005

Citing 2
Cited 1

A survey of design techniques for system-level dynamic power management

IEEE Transactions on Very Large Scale Integration (VLSI) Systems - Special section on low-power electronics and design
Reinforcement Learning

Reinforcement Learning

Adaptive Power Management Based on Reinforcement Learning for Embedded System

IEA/AIE '08 Proceedings of the 21st international conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems: New Frontiers in Applied Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

When applying Dynamic Power Management (DPM) technique to pervasively deployed embedded systems, the technique needs to be very efficient so that it is feasible to implement the technique on low end processor and tight-budget memory. Furthermore, it should have the capability to track time varying behavior rapidly because the time varying is an inherent characteristic of real world system. Existing methods, which are usually model-based, may not satisfy the aforementioned requirements. In this paper, we propose a model-free DPM technique based on Q-Learning. Q-DPM is much more efficient because it removes the overhead of parameter estimator and mode-switch controller. Furthermore, its policy optimization is performed via consecutive online trialing, which also leads to very rapid response to time varying behavior.