MILEPOST GCC: machine learning based research compiler (2008)
Fursin, Grigori, Miranda, Cupertino, Temam, Olivier, Namolaru, Mircea, Yom-Tov, Elad, Zaks, Ayal, ...
Tuning hardwired compiler optimizations for rapidly evolving hardware makes porting an optimizing compiler for each new platform extremely challenging. Our radical approach is to develop a modular,...
MILEPOST GCC: machine learning based research compiler (2008)
Fursin, Grigori, Miranda, Cupertino, Temam, Olivier, Namolaru, Mircea, Yom-Tov, Elad, Zaks, Ayal, ...
Tuning hardwired compiler optimizations for rapidly evolving hardware makes porting an optimizing compiler for each new platform extremely challenging. Our radical approach is to develop a modular,...
Kernel Multi-task Learning using Task-specific Features. (2007)
Bonilla, Edwin, Agakov, Felix, Williams, Christopher
In this paper we are concerned with multitask learning when task-specific features are available. We describe two ways of achieving this using Gaussian process predictors: in the first method, the...
Cavazos, John, Dubach, Christophe, Agakov, Felix, Bonilla, Edwin, O'Boyle, Michael, Fursin, Grigori, ...
Developing an optimizing compiler for a newly proposed architecture is extremely difficult when there is only a simulator of the machine available. Designing such a compiler requires running many...
Predictive Search Distributions (2006)
Bonilla, Edwin, Williams, Christopher, Agakov, Felix, Cavazos, John, Thomson, John, O'Boyle, Michael
Estimation of Distribution Algorithms (EDAs) are a popular approach to learn a probability distribution over the "good" solutions to a combinatorial optimization problem. Here we consider the case...
Using Machine Learning to Focus Iterative Optimization (2006)
Agakov, Felix, Bonilla, Edwin, Cavazos, John, Franke, Bjoern, Fursin, Grigori, O'Boyle, Michael, ...
Iterative compiler optimization has been shown to outperform static approaches. This, however, is at the cost of large numbers of evaluations of the program. This paper develops a new methodology to...