STAFF

Online learning with stabilized adaptive forgetting factor and feature selection algorithm

Ujjwal Gupta, Manoj Babu, Raid Ayoub, Michael Kishinevsky, Francesco Paterna, Umit Ogras

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

2 Citations (Scopus)

Abstract

Dynamic resource management techniques rely on power consumption and performance models to optimize the operating frequency and utilization of processing elements, such as CPU and GPU. Despite the importance of these decisions, many existing approaches rely on fixed power and performance models that are learned online. However, o.ine models cannot guarantee accuracy when workloads differ signifucantly from the training available at design time. This paper presents an online learning framework (STAFF) that constructs adaptive run-time models for stationary and non-stationary workloads. STAFF is the first framework that (1) guarantees stability while quickly adapting to workload changes, (2) performs online feature selection with linear complexity, and (3) adapts to new model coe.cients by employing adaptively varying forgetting factor, all at the same time. Experiments on an Intelr CoreTM i5 6th generation platform demonstrate up to 6. improvement in the performance prediction accuracy compared to existing techniques.

Original languageEnglish (US)
Title of host publicationProceedings of the 55th Annual Design Automation Conference, DAC 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
VolumePart F137710
ISBN (Print)9781450357005
DOIs
StatePublished - Jun 24 2018
Event55th Annual Design Automation Conference, DAC 2018 - San Francisco, United States
Duration: Jun 24 2018Jun 29 2018

Other

Other55th Annual Design Automation Conference, DAC 2018
CountryUnited States
CitySan Francisco
Period6/24/186/29/18

Fingerprint

Online Learning
Feature Selection
Workload
Feature extraction
Performance Model
Linear Complexity
Performance Prediction
Resource Management
Power Consumption
Optimise
Model
Program processors
Electric power utilization
Demonstrate
Experiment
Processing
Framework
Experiments

ASJC Scopus subject areas

  • Computer Science Applications
  • Control and Systems Engineering
  • Electrical and Electronic Engineering
  • Modeling and Simulation

Cite this

Gupta, U., Babu, M., Ayoub, R., Kishinevsky, M., Paterna, F., & Ogras, U. (2018). STAFF: Online learning with stabilized adaptive forgetting factor and feature selection algorithm. In Proceedings of the 55th Annual Design Automation Conference, DAC 2018 (Vol. Part F137710). [a177] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1145/3195970.3196122

STAFF : Online learning with stabilized adaptive forgetting factor and feature selection algorithm. / Gupta, Ujjwal; Babu, Manoj; Ayoub, Raid; Kishinevsky, Michael; Paterna, Francesco; Ogras, Umit.

Proceedings of the 55th Annual Design Automation Conference, DAC 2018. Vol. Part F137710 Institute of Electrical and Electronics Engineers Inc., 2018. a177.

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

Gupta, U, Babu, M, Ayoub, R, Kishinevsky, M, Paterna, F & Ogras, U 2018, STAFF: Online learning with stabilized adaptive forgetting factor and feature selection algorithm. in Proceedings of the 55th Annual Design Automation Conference, DAC 2018. vol. Part F137710, a177, Institute of Electrical and Electronics Engineers Inc., 55th Annual Design Automation Conference, DAC 2018, San Francisco, United States, 6/24/18. https://doi.org/10.1145/3195970.3196122
Gupta U, Babu M, Ayoub R, Kishinevsky M, Paterna F, Ogras U. STAFF: Online learning with stabilized adaptive forgetting factor and feature selection algorithm. In Proceedings of the 55th Annual Design Automation Conference, DAC 2018. Vol. Part F137710. Institute of Electrical and Electronics Engineers Inc. 2018. a177 https://doi.org/10.1145/3195970.3196122
Gupta, Ujjwal ; Babu, Manoj ; Ayoub, Raid ; Kishinevsky, Michael ; Paterna, Francesco ; Ogras, Umit. / STAFF : Online learning with stabilized adaptive forgetting factor and feature selection algorithm. Proceedings of the 55th Annual Design Automation Conference, DAC 2018. Vol. Part F137710 Institute of Electrical and Electronics Engineers Inc., 2018.
@inproceedings{366f0247cff64ff982317fdbb4c6ed76,
title = "STAFF: Online learning with stabilized adaptive forgetting factor and feature selection algorithm",
abstract = "Dynamic resource management techniques rely on power consumption and performance models to optimize the operating frequency and utilization of processing elements, such as CPU and GPU. Despite the importance of these decisions, many existing approaches rely on fixed power and performance models that are learned online. However, o.ine models cannot guarantee accuracy when workloads differ signifucantly from the training available at design time. This paper presents an online learning framework (STAFF) that constructs adaptive run-time models for stationary and non-stationary workloads. STAFF is the first framework that (1) guarantees stability while quickly adapting to workload changes, (2) performs online feature selection with linear complexity, and (3) adapts to new model coe.cients by employing adaptively varying forgetting factor, all at the same time. Experiments on an Intelr CoreTM i5 6th generation platform demonstrate up to 6. improvement in the performance prediction accuracy compared to existing techniques.",
author = "Ujjwal Gupta and Manoj Babu and Raid Ayoub and Michael Kishinevsky and Francesco Paterna and Umit Ogras",
year = "2018",
month = "6",
day = "24",
doi = "10.1145/3195970.3196122",
language = "English (US)",
isbn = "9781450357005",
volume = "Part F137710",
booktitle = "Proceedings of the 55th Annual Design Automation Conference, DAC 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - STAFF

T2 - Online learning with stabilized adaptive forgetting factor and feature selection algorithm

AU - Gupta, Ujjwal

AU - Babu, Manoj

AU - Ayoub, Raid

AU - Kishinevsky, Michael

AU - Paterna, Francesco

AU - Ogras, Umit

PY - 2018/6/24

Y1 - 2018/6/24

N2 - Dynamic resource management techniques rely on power consumption and performance models to optimize the operating frequency and utilization of processing elements, such as CPU and GPU. Despite the importance of these decisions, many existing approaches rely on fixed power and performance models that are learned online. However, o.ine models cannot guarantee accuracy when workloads differ signifucantly from the training available at design time. This paper presents an online learning framework (STAFF) that constructs adaptive run-time models for stationary and non-stationary workloads. STAFF is the first framework that (1) guarantees stability while quickly adapting to workload changes, (2) performs online feature selection with linear complexity, and (3) adapts to new model coe.cients by employing adaptively varying forgetting factor, all at the same time. Experiments on an Intelr CoreTM i5 6th generation platform demonstrate up to 6. improvement in the performance prediction accuracy compared to existing techniques.

AB - Dynamic resource management techniques rely on power consumption and performance models to optimize the operating frequency and utilization of processing elements, such as CPU and GPU. Despite the importance of these decisions, many existing approaches rely on fixed power and performance models that are learned online. However, o.ine models cannot guarantee accuracy when workloads differ signifucantly from the training available at design time. This paper presents an online learning framework (STAFF) that constructs adaptive run-time models for stationary and non-stationary workloads. STAFF is the first framework that (1) guarantees stability while quickly adapting to workload changes, (2) performs online feature selection with linear complexity, and (3) adapts to new model coe.cients by employing adaptively varying forgetting factor, all at the same time. Experiments on an Intelr CoreTM i5 6th generation platform demonstrate up to 6. improvement in the performance prediction accuracy compared to existing techniques.

UR - http://www.scopus.com/inward/record.url?scp=85053682642&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85053682642&partnerID=8YFLogxK

U2 - 10.1145/3195970.3196122

DO - 10.1145/3195970.3196122

M3 - Conference contribution

SN - 9781450357005

VL - Part F137710

BT - Proceedings of the 55th Annual Design Automation Conference, DAC 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -