Can a generative adversarial network (GAN) be used for data augmentation (i.e. to generate synthetic examples that are added to a dataset) for data that is tabular/vectorized (i.e. not an image)? Are there any public implementations of methods to this effect.
I am aware of the following papers on images: https://arxiv.org/abs/1711.04340 https://arxiv.org/abs/1801.05401 https://arxiv.org/abs/1803.01229
GANs are primarily used for data augmentation. If you have 1D signals you could use MLP or 1D convolutions. Hope those links will help: http://www.rricard.me/machine/learning/generative/adversarial/networks/keras/tensorflow/2017/04/05/gans-part2.html https://github.com/timzhang642/GAN-1D-Gaussian-Distribution
Of course, you can generate some new data as data augmentation, check out
Take look at review two recent papers https://towardsdatascience.com/review-of-gans-for-tabular-data-a30a2199342
