Title :
Adaptive selection of helper-objectives for test case generation
Author :
Buzdalov, Maxim ; Buzdalova, Arina
Author_Institution :
St. Petersburg Nat. Res. Univ. of Inf. Technol., Mech. & Opt., St. Petersburg, Russia
Abstract :
In this paper a method of adaptive selection of helper-objectives in evolutionary algorithms, which was previously applied to model problems only, is applied to generation of test cases for programming challenge tasks. The method is based on reinforcement learning. Experiments show that the proposed method performs equally well compared to the best helper-objectives selected by hand.
Keywords :
evolutionary computation; learning (artificial intelligence); adaptive helper-objectives selection; evolutionary algorithms; reinforcement learning; test case generation; Evolutionary computation; Genetic algorithms; Learning (artificial intelligence); Optimization; Programming; Sociology; Statistics;
Conference_Titel :
Evolutionary Computation (CEC), 2013 IEEE Congress on
Conference_Location :
Cancun
Print_ISBN :
978-1-4799-0453-2
Electronic_ISBN :
978-1-4799-0452-5
DOI :
10.1109/CEC.2013.6557836
