![](bg9.png)
et al., 2020) and an approach using an artificial neural
network-based (Spieker et al., 2017).
We used a XCSF-based agent and employed a
simple heuristic: a test case of high value should have
a high priority. Hence we used XCSF to approximate
a state-value function V (·) and interpreted the approx-
imated values as actions.
We benchmarked our agent on three different data
sets using three reward functions. In our comparison,
XCSF was in 8 out of 9 cases superior to the neural
network. For a single combination of reward func-
tion and data set XCSF was inferior. However, if the
best combinations of reward functions and agent are
considered on the data set then both approaches are
equal.
Our experiments showed that the continuous out-
put leads to a performance boost compared to XCS as
XCSF was in all nine cases considered either superior
or had an equivalent performance. Thus we recom-
mend to use XCSF for ATCS.
REFERENCES
Aggarwal, C. (2020). Linear Algebra and Optimization for
Machine Learning: A Textbook.
Butz, M. V. and Wilson, S. W. (2001). ”An Algorithmic
Description of XCS”. In Luca Lanzi, P., Stolzmann,
W., and Wilson, S. W., editors, Advances in Learn-
ing Classifier Systems, pages 253–272, Berlin, Hei-
delberg. Springer Berlin Heidelberg.
Dai, Y., Xie, M., Poh, K., and Yang, B. (2003). ”Optimal
testing-resource allocation with genetic algorithm for
modular software systems”. Journal of Systems and
Software, 66(1):47 – 55.
Di Nardo, D., Alshahwan, N., Briand, L., and Labiche, Y.
(2015). Coverage-based regression test case selection,
minimization and prioritization: a case study on an
industrial system. Software Testing, Verification and
Reliability, 25(4):371–396.
Dustin, E., Rashka, J., and Paul, J. (1999). Automated Soft-
ware Testing: Introduction, Management, and Perfor-
mance. Addison-Wesley Longman Publishing Co.,
Inc., USA.
Epitropakis, M. G., Yoo, S., Harman, M., and Burke, E. K.
(2015). Empirical evaluation of pareto efficient multi-
objective regression test case prioritisation. In ISSTA
2015.
Gligoric, M., Eloussi, L., and Marinov, D. (2015). Ekstazi:
Lightweight Test Selection. In 2015 IEEE/ACM 37th
IEEE International Conference on Software Engineer-
ing, volume 2, pages 713–716.
Haga, H. and Suehiro, A. (2012). Automatic test case gen-
eration based on genetic algorithm and mutation anal-
ysis. In 2012 IEEE International Conference on Con-
trol System, Computing and Engineering, pages 119–
123.
Haghighatkhah, A. (2020). Test case prioritization using
build history and test distances: an approach for im-
proving automotive regression testing in continuous
integration environments.
Hamid, O. H. and Braun, J. (2019). Reinforcement Learn-
ing and Attractor Neural Network Models of Associa-
tive Learning, pages 327–349. Springer International
Publishing, Cham.
Heider, M., P
¨
atzel, D., and H
¨
ahner, J. (2020a). Towards
a Pittsburgh-Style LCS for Learning Manufacturing
Machinery Parametrizations. In Proceedings of the
2020 Genetic and Evolutionary Computation Confer-
ence Companion, GECCO ’20, page 127–128, New
York, NY, USA. Association for Computing Machin-
ery.
Heider, M., P
¨
atzel, D., and H
¨
ahner, J. (2020b). SupRB: A
Supervised Rule-based Learning System for Continu-
ous Problems.
Jia, Y. and Harman, M. (2008). Constructing Subtle Faults
Using Higher Order Mutation Testing. In 2008 Eighth
IEEE International Working Conference on Source
Code Analysis and Manipulation, pages 249 – 258.
Jung-Min Kim and Porter, A. (2002). A history-based test
prioritization technique for regression testing in re-
source constrained environments. In Proceedings of
the 24th International Conference on Software Engi-
neering. ICSE 2002, pages 119–129.
Kwon, J., Ko, I., Rothermel, G., and Staats, M. (2014). Test
Case Prioritization Based on Information Retrieval
Concepts. In 2014 21st Asia-Pacific Software Engi-
neering Conference, volume 1, pages 19–26.
Land, K., Cha, S., and Vogel-Heuser, B. (2019). An
Approach to Efficient Test Scheduling for Auto-
mated Production Systems. 2019 IEEE 17th Interna-
tional Conference on Industrial Informatics (INDIN),
1:449–454.
Lanzi, P. L. and Loiacono, D. (2010). ”Speeding Up Match-
ing in Learning Classifier Systems Using CUDA”. In
Bacardit, J., Browne, W., Drugowitsch, J., Bernad
´
o-
Mansilla, E., and Butz, M. V., editors, Learning
Classifier Systems, pages 1–20, Berlin, Heidelberg.
Springer Berlin Heidelberg.
Lanzi, P. L., Loiacono, D., Wilson, S. W., and Goldberg,
D. E. (2005). Extending xcsf beyond linear approxi-
mation. In Proceedings of the 7th Annual Conference
on Genetic and Evolutionary Computation, GECCO
’05, page 1827–1834, New York, NY, USA. Associa-
tion for Computing Machinery.
Lee, P., Teng, Y., and Hsiao, T.-C. (2012). XCSF for Predic-
tion on Emotion Induced by Image Based on Dimen-
sional Theory of Emotion. In Proceedings of the 14th
Annual Conference Companion on Genetic and Evo-
lutionary Computation, GECCO ’12, page 375–382,
New York, NY, USA. Association for Computing Ma-
chinery.
Marijan, D., Gotlieb, A., and Sen, S. (2013). Test Case Pri-
oritization for Continuous Regression Testing: An In-
dustrial Case Study. In 2013 IEEE International Con-
ference on Software Maintenance, pages 540–543.
Mirarab, S., Akhlaghi, S., and Tahvildari, L. (2012). Size-
Constrained Regression Test Case Selection Using
XCSF for Automatic Test Case Prioritization
57