6 min read1 hour ago

–

The applications of deep learning are widespread; it helps in building natural language processing models, fraud detection, and more. Deep learning is also the source of some of the complex techniques, one of which is a generative adversarial network and a deep convolution generative adversarial network(DCGAN).

A generative adversarial network is a way to implement generative modelling using deep learning techniques. A generative model is basically a model that predicts the next value, like the next word in a sentence or more. Let’s learn the difference between discriminative modelling and generative modelling to understand generative modelling better.

Discriminative modeling VS Generative Modeling

Both the discriminative model and the generative model could be used for the classification problem, and both could also be the supervised problem. The difference lies in the way they separate positives from negatives. The discriminative model draws the boundary between the two types of data, whereas the generative model will draw two classes that are distinct from one another.

Press enter or click to view image in full size

Discriminative Model VS Generative Model

The generative model uses an unsupervised learning approach, which means the output that is generated by the model is…