Hardware architecture of reinforcement learning scheme for dynamic power management in embedded systems

Authors:
Viswanathan Lakshmi Prabha;Elwin Chandra Monie
Affiliations:
Department of Electronics and Communication Engineering, Government College of Technology, Coimbatore, Tamil Nadu, India;Thanthai Periyar Government Institute of Technology, Vellore, Tamil Nadu, India
Venue:
EURASIP Journal on Embedded Systems
Year:
2007

Citing 7
Cited 4

Software controlled power management

CODES '99 Proceedings of the seventh international workshop on Hardware/software codesign
Dynamic power management based on continuous-time Markov decision processes

Proceedings of the 36th annual ACM/IEEE Design Automation Conference
Dynamic power management using adaptive learning tree

ICCAD '99 Proceedings of the 1999 IEEE/ACM international conference on Computer-aided design
Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
Comparing System-Level Power Management Policies

IEEE Design & Test
A Model Checking Approach to Evaluating System Level Dynamic Power Management Policies for Embedded Systems

HLDVT '01 Proceedings of the Sixth IEEE International High-Level Design Validation and Test Workshop (HLDVT'01)
Policy optimization for dynamic power management

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems

Adaptive power management using reinforcement learning

Proceedings of the 2009 International Conference on Computer-Aided Design
Enhanced Q-learning algorithm for dynamic power management with performance constraint

Proceedings of the Conference on Design, Automation and Test in Europe
An adaptive hybrid dynamic power management algorithm for mobile devices

Computer Networks: The International Journal of Computer and Telecommunications Networking
Online learning of timeout policies for dynamic power management

ACM Transactions on Embedded Computing Systems (TECS)

Quantified Score

Hi-index	0.00

Visualization

Abstract

Dynamic power management (DPM) is a technique to reduce power consumption of electronic systems by selectively shutting down idle components. In this paper, a novel and nontrivial enhancement of conventional reinforcement learning (RL) is adopted to choose the optimal policy out of the existing DPM policies. A hardware architecture evolved from the VHDL model of Temporal Difference RL algorithm is proposed in this paper, which can suggest the winner policy to be adopted for any given workload to achieve power savings. The effectiveness of this approach is also demonstrated by an event-driven simulator, which is designed using JAVA for power-manageable embedded devices. The results show that RL applied to DPM can lead up to 28% power savings.