@inproceedings{discovery10094843, address = {Tallinn, Estonia}, note = {This version is the author accepted manuscript. For information on re-use, please refer to the publisher's terms and conditions.}, volume = {11664}, editor = {S Nejati and G Gay}, pages = {27--41}, booktitle = {Lecture Notes in Computer Science}, month = {August}, journal = {Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)}, publisher = {Springer}, title = {Constructing Search Spaces for Search-Based Software Testing Using Neural Networks}, year = {2019}, author = {Joffe, L and Clark, D}, abstract = {A central requirement for any Search-Based Software Testing (SBST) technique is a convenient and meaningful fitness landscape. Whether one follows a targeted or a diversification driven strategy, a search landscape needs to be large, continuous, easy to construct and representative of the underlying property of interest. Constructing such a landscape is not a trivial task often requiring a significant manual effort by an expert. We present an approach for constructing meaningful and convenient fitness landscapes using neural networks (NN) - for targeted and diversification strategies alike. We suggest that output of an NN predictor can be interpreted as a fitness for a targeted strategy. The NN is trained on a corpus of execution traces and various properties of interest, prior to searching. During search, the trained NN is queried to predict an estimate of a property given an execution trace. The outputs of the NN form a convenient search space which is strongly representative of a number of properties. We believe that such a search space can be readily used for driving a search towards specific properties of interest. For a diversification strategy, we propose the use of an autoencoder; a mechanism for compacting data into an n-dimensional "latent" space. In it, datapoints are arranged according to the similarity of their salient features. We show that a latent space of execution traces possesses characteristics of a convenient search landscape: it is continuous, large and crucially, it defines a notion of similarity to arbitrary observations.}, url = {https://doi.org/10.1007/978-3-030-27455-9\%5f3}, keywords = {Search-Based Software Testing . Software Engineering . Fitness Function . Machine Learning . Neural Networks} }