Autoencoder based generators of semi-artificial dataKlančar, Jaka (Avtor) Robnik Šikonja, Marko (Mentor) autoencodersdata generatorsneural networkssemi-artifical datavariational autoencodersThe goal of the thesis is to alleviate the problem of insufficient data available for data analysis or machine learning. We developed a generator of semi-artificial data based on autoencoders. We implemented dynamic autoencoders without any predefined structure, as we wanted that our solution is general and may therefore be used on any data set. Results showed that autoencoder based generators work better than variational autoencoders. The generators perform best on data sets with a small number of mixed attributes and balanced classes. They perform better if more training instances are available. Results additionally show that grid search significantly improves the performance and that it is possible to predict a good set of parameters for each data set.20182018-07-26 15:51:47Magistrsko delo/naloga102239VisID: 21103sl