The generative network's training objective is to increase the error rate of the discriminative network (i.e., "fool" the discriminator network by producing novel candidates that the discriminator thinks are not synthesized (are part of the true data distribution)). Typically, the generative network learns to map from a latent space to a data distribution of interest, while the discriminative network distinguishes candidates produced by the generator from the true data distribution. The contest operates in terms of data distributions. The generative network generates candidates while the discriminative network evaluates them.
2.5 Concerns about malicious applications.This enables the model to learn in an unsupervised manner. This basically means that the generator is not trained to minimize the distance to a specific image, but rather to fool the discriminator. The core idea of a GAN is based on the "indirect" training through the discriminator, another neural network that is able to tell how much an input is "realistic", which itself is also being updated dynamically. Though originally proposed as a form of generative model for unsupervised learning, GANs have also proved useful for semi-supervised learning, fully supervised learning, and reinforcement learning. For example, a GAN trained on photographs can generate new photographs that look at least superficially authentic to human observers, having many realistic characteristics. Given a training set, this technique learns to generate new data with the same statistics as the training set. Two neural networks contest with each other in a game (in the form of a zero-sum game, where one agent's gain is another agent's loss). A generative adversarial network ( GAN) is a class of machine learning frameworks designed by Ian Goodfellow and his colleagues in June 2014.
0 kommentar(er)
