Deep convolutional generative adversarial networks

The deep convolutional generative adversarial networks (DCGAN) are introduced in the paper: Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks, by A. Radford, L. Metz, and S. Chintala, arXiv: 1511.06434, 2015. The generator uses a 100-dimensional, uniform distribution space, Z, which is then projected into a smaller space by a series of vis-a-vis convolution operations. An example is shown in the following figure:

A DCGAN generator can be described by the following Keras code; it is also described by one implementation, available at: ...

Get Deep Learning with Keras now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.