DocumentCode :
638769
Title :
Savant: Automatic parallelization of a scheduling heuristic with machine learning
Author :
Pinel, Frederic ; Bouvry, Pascal ; Dorronsoro, Bernabe ; Khan, Samee U.
Author_Institution :
FSTC/CSC/ILIAS, Univ. of Luxembourg, Luxembourg City, Luxembourg
fYear :
2013
fDate :
12-14 Aug. 2013
Firstpage :
52
Lastpage :
57
Abstract :
This paper investigates the automatic parallelization of a heuristic for an NP-complete problem, with machine learning. The objective is to automatically design a new concurrent algorithm that finds solutions of comparable quality to the original heuristic. Our approach, called Savant, is inspired from the Savant syndrome. Its concurrency model is based on map-reduce. The approach is evaluated with the well-known Min-Min heuristic. Simulation results on two problem sizes are promising, the produced algorithm is able to find solutions of comparable quality.
Keywords :
learning (artificial intelligence); optimisation; parallel algorithms; parallelising compilers; scheduling; NP-complete problem; Savant syndrome; automatic parallelization; concurrency model; concurrent algorithm; machine learning; map-reduce; min-min heuristic; scheduling heuristic; Power capacitors; Conversion from sequential to parallel forms; Parallelism and concurrency; Pattern matching;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Nature and Biologically Inspired Computing (NaBIC), 2013 World Congress on
Conference_Location :
Fargo, ND
Print_ISBN :
978-1-4799-1414-2
Type :
conf
DOI :
10.1109/NaBIC.2013.6617837
Filename :
6617837
Link To Document :
بازگشت