Accurate and efficient processor performance prediction via regression tree based modeling
Citations
276 citations
Cites background or methods from "Accurate and efficient processor pe..."
...For the improvement of global accuracy, a distance term d, which plays a role of global exploration, is often required to discard some points that are close to each other (Hendrickx and Dhaene 2005; Li et al. 2009; Eason and Cremaschi 2014)....
[...]
...With no constraints, the new points tend to cluster in regions with large prediction difference (Li et al. 2009)....
[...]
...Li et al. (2009) trained the multiple additive regression trees (MART) model on the sample set 20 times with different random seeds, and suggested using the coefficient of variance σ ̂=μ (where μ represents the mean) instead of the QBC variance to better rank the candidate points....
[...]
93 citations
37 citations
Cites methods from "Accurate and efficient processor pe..."
...The selection of a specific type of RSM (model selection) has been addressed by means of data mining [3], genetically evolved neural networks [28], and genetically evolved linear models [29]....
[...]
...For these reasons, analytical models [or response surface models (RSMs)] are increasingly used to approximate the profiling information obtained with computer simulations by effectively replacing it during optimization [3], [4]....
[...]
37 citations
32 citations
References
21,694 citations
17,764 citations
"Accurate and efficient processor pe..." refers background in this paper
...[6] show that using small values of gradient descent step size always lead to better prediction performance....
[...]
16,974 citations
14,825 citations