How can generative adversarial networks learn real-life distributions easily

Figure 5: our experiments verify that after training a 4-layer GAN, the hidden layers become very sparse compared to their random initializations.

Figure 7: the first hidden layer in the discriminator learns edge-color detectors, while the output layer of the generator also learns edge-color features.

Figure 8: the GAN framework can simulate supervised learning to learn 1-hidden-layer network from inputs \(S_{l-1}^*\) to outputs \(S_l^*\).

Leave a Comment