Generative Adversarial Nets: Can we generate a new dataset based on only one training set?
This work addresses the need for generating novel datasets beyond the training distribution, which is incremental as it builds on GANs by adding control over distribution divergence.
The paper tackles the problem of generating a new dataset with a different distribution from the training set, controlling the Jensen-Shannon divergence between distributions by a target δ, and demonstrates this with an application to generating new kinds of rice with similar characteristics to good rice.
A generative adversarial network (GAN) is a class of machine learning frameworks designed by Goodfellow et al. in 2014. In the GAN framework, the generative model is pitted against an adversary: a discriminative model that learns to determine whether a sample is from the model distribution or the data distribution. GAN generates new samples from the same distribution as the training set. In this work, we aim to generate a new dataset that has a different distribution from the training set. In addition, the Jensen-Shannon divergence between the distributions of the generative and training datasets can be controlled by some target $δ\in [0, 1]$. Our work is motivated by applications in generating new kinds of rice that have similar characteristics as good rice.