Enhanced Q-learning algorithm for dynamic power management with performance constraint

Authors:
Wei Liu;Ying Tan;Qinru Qiu
Affiliations:
Binghamton University, State University of New York, Binghamton, New York;Binghamton University, State University of New York, Binghamton, New York;Binghamton University, State University of New York, Binghamton, New York
Venue:
Proceedings of the Conference on Design, Automation and Test in Europe
Year:
2010

Citing 6
Cited 2

Dynamic power management using machine learning

Proceedings of the 2006 IEEE/ACM international conference on Computer-aided design
Stochastic modeling and optimization for robust power management in a partially observable system

Proceedings of the conference on Design, automation and test in Europe
Dynamic power management under uncertain information

Proceedings of the conference on Design, automation and test in Europe
Hardware architecture of reinforcement learning scheme for dynamic power management in embedded systems

EURASIP Journal on Embedded Systems
Policy optimization for dynamic power management

IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Event-driven power management

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

Achieving autonomous power management using reinforcement learning

ACM Transactions on Design Automation of Electronic Systems (TODAES)
A survey of multi-objective sequential decision-making

Journal of Artificial Intelligence Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a novel power management techniques based on enhanced Q-learning algorithms. By exploiting the submodularity and monotonic structure in the cost function of a power management system, the enhanced Q-learning algorithm is capable of exploring ideal trade-offs in the power-performance design space and converging to a better power management policy. We further propose a linear adaption algorithm that adapts the Lagrangian multiplier λ to search for the power management policy that minimizes the power consumption while delivering the exact required performance. Experimental results show that, comparing to the existing expert-based power management, the proposed Q-learning based power management achieves up to 30% and 60% reduction in power saving for synthetic workload and real workload, respectively while in average maintain a performance within 7% variation of the given constraint.