leo-p
commented
7 years ago Summary:
- Dataset: MNIST and MIRFLICKR.
- Objective: In an unconditional GAN it's not possible to control the mode of the data being generated which is what this paper tries to accomplish using the label data (but it can be generalized to any kind of conditional data).
Inner workings:
Changes the loss to the conditional loss:
For implementation the only thing needed is to feed the label data to both the discriminator and generator:
Results:
Interesting at the time but not surprising now. There's not much more to the paper than what is in the summary.
https://arxiv.org/pdf/1411.1784.pdf
Generative Adversarial Nets were recently introduced as a novel way to train generative models. In this work we introduce the conditional version of generative adversarial nets, which can be constructed by simply feeding the data, y, we wish to condition on to both the generator and discriminator. We show that this model can generate MNIST digits conditioned on class labels. We also illustrate how this model could be used to learn a multi-modal model, and provide preliminary examples of an application to image tagging in which we demonstrate how this approach can generate descriptive tags which are not part of training labels.