GANS for Sequences of Discrete Elements with the Gumbel-softmax Distribution

Generative Adversarial Networks (GAN) have limitations when the goal is to sequences of discrete elements. The reason for this is that samples a distribution on discrete objects such as the multinomial are not with respect to the distribution parameters. This problem can be by using the Gumbel-softmax distribution, which is a continuous to a multinomial distribution parameterized in terms of the function. In this work, we evaluate the performance of GANs based on neural networks with Gumbel-softmax output distributions in the task generating sequences of discrete elements.