TY - GEN
T1 - Accuracy analysis of machine learning-based performance modeling for microprocessors
AU - Tanaka, Yoshihiro
AU - Oka, Keitaro
AU - Ono, Takatsugu
AU - Inoue, Koji
N1 - Funding Information:
This research was supported in part by JSPS KAKENHI Grant Number 26540022 and the Japan Science and Technology Agency (JST) CREST program.
Publisher Copyright:
© 2016 IEEE.
PY - 2016/7/21
Y1 - 2016/7/21
N2 - This paper analyzes accuracy of performance models generated by machine learning-based empirical modeling methodology. Although the accuracy strongly depends on the quality of learning procedure, it is not clear what kind of learning algorithms and training data set (or feature) should be used. This paper inclusively explores the learning space of processor performance modeling as a case study. We focus on static architectural parameters as training data set such as cache size and clock frequency. Experimental results show that a tree-based non-linear regression modeling is superior to a stepwise linear regression modeling. Another observation is that clock frequency is the most important feature to improve prediction accuracy.
AB - This paper analyzes accuracy of performance models generated by machine learning-based empirical modeling methodology. Although the accuracy strongly depends on the quality of learning procedure, it is not clear what kind of learning algorithms and training data set (or feature) should be used. This paper inclusively explores the learning space of processor performance modeling as a case study. We focus on static architectural parameters as training data set such as cache size and clock frequency. Experimental results show that a tree-based non-linear regression modeling is superior to a stepwise linear regression modeling. Another observation is that clock frequency is the most important feature to improve prediction accuracy.
UR - http://www.scopus.com/inward/record.url?scp=84991810169&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84991810169&partnerID=8YFLogxK
U2 - 10.1109/JEC-ECC.2016.7518973
DO - 10.1109/JEC-ECC.2016.7518973
M3 - Conference contribution
AN - SCOPUS:84991810169
T3 - Proceedings of the 2016 4th International Japan-Egypt Conference on Electronic, Communication and Computers, JEC-ECC 2016
SP - 83
EP - 86
BT - Proceedings of the 2016 4th International Japan-Egypt Conference on Electronic, Communication and Computers, JEC-ECC 2016
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 4th International Japan-Egypt Conference on Electronic, Communication and Computers, JEC-ECC 2016
Y2 - 31 May 2016 through 2 June 2016
ER -