Abstract: Generative Adversarial Networks, or GAN for short, is a productive modeling model using in-depth learning methods, such as convolutional neural networks.
In recent times, supervised learning with convolutional networks (CNNs) has seen huge adoption in computer vision applications. Comparatively, unsupervised learning with CNNs has received less attention.
We introduce a class of CNNs called deep convolutional generative adversarial networks (DCGANs), that have certain constraints, and demonstrate that they are a strong candidate for unsupervised learning.
Training on various image datasets, such as anime we show convincing evidence that our deep convolutional adversarial pair learns a hierarchy of representations from both the generator and discriminator. Additionally, we use the learned features for novel tasks - demonstrating their applicability to be used for various purposes such as advertising.
|
DOI:
10.17148/IJARCCE.2022.115209
