Adaptive power management using reinforcement learning

Ying Tan, Wei Liu, Qinru Qiu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

80 Scopus citations

Abstract

System level power management must consider the uncertainty and variability that comes from the environment, the application and the hardware. A robust power management technique must be able to learn the optimal decision from past history and improve itself as the environment changes. This paper presents a novel online power management technique based on model-free constrained reinforcement learning (RL). It learns the best power management policy that gives the minimum power consumption for a given performance constraint without any prior information of workload. Compared with existing machine learning based power management techniques, the RL based learning is capable of exploring the trade-off in the power-performance design space and converging to a better power management policy. Experimental results show that the proposed RL based power management achieves 24% and 3% reduction in power and latency respectively comparing to the existing expert based power management.

Original languageEnglish (US)
Title of host publicationProceedings of the 2009 IEEE/ACM International Conference on Computer-Aided Design - Digest of Technical Papers, ICCAD 2009
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages461-467
Number of pages7
ISBN (Print)9781605588001
DOIs
StatePublished - Jan 1 2009
Externally publishedYes
Event2009 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2009 - San Jose, CA, United States
Duration: Nov 2 2009Nov 5 2009

Publication series

NameIEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD
ISSN (Print)1092-3152

Other

Other2009 IEEE/ACM International Conference on Computer-Aided Design, ICCAD 2009
CountryUnited States
CitySan Jose, CA
Period11/2/0911/5/09

Keywords

  • Model-free
  • Power management
  • Q-learning
  • Reinforcement learning

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design

Fingerprint Dive into the research topics of 'Adaptive power management using reinforcement learning'. Together they form a unique fingerprint.

  • Cite this

    Tan, Y., Liu, W., & Qiu, Q. (2009). Adaptive power management using reinforcement learning. In Proceedings of the 2009 IEEE/ACM International Conference on Computer-Aided Design - Digest of Technical Papers, ICCAD 2009 (pp. 461-467). [5361254] (IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1145/1687399.1687486