This technique has been applied to several loop optimizations, including loop interchange, loop tiling, and software pipelining and appears to be quite promising.This article is about flash-based, DRAM-based, and other solid-state storage.Once saturated, the intermediate representation encodes multiple optimized versions of the input program.At this point, a profitability heuristic picks the final optimized program from the various programs represented in the saturated representation.cross validation as part of the model fitting procedure.That means that the fitting including the fitting of the hyper-parameters (this is where the inner cross validation hides) is just like any other model esitmation routine.SVMs with different kernels, trained with possibly different features, depending on the grid search). It looks to me that selecting the best model out of those $K$ winning models would not be a fair comparison since each model was trained and tested on different parts of the dataset. Also I have read threads discussing how nested model selection is useful for analyzing the learning procedure.
The quality and robustness of the Co Sy system and the compilers it generates are equally important.
And as the focus in producing compilers is moved to creation of the processor description, optimizing compilers can be available as soon as architecture specifications are stable.
Compilers built with the Co Sy compiler development system are inherently of high quality and performance.
Published at POPL 2009: [pdf] [bibtex] [pptx] [mp4] Journal Version in LMCS 2011: ar Xiv pdf [bibtex] Authors: Ross Tate, Michael Stepp, Zachary Tatlock, Sorin Lerner Conversion Algorithms Technical Report: [pdf] [bibtex] CAV Tools 2011 on application to translation validation for LLVM: [pdf] [bibtex] Theses: Ross Tate's [pdf] and Michael Stepp's [pdf] We have followed up this research with a technique for automatically learning optimizations.
In particular, this technique can be used to mitigate the cost of equality saturation and global profitability heuristics.