• Daliboyina Nikitha
• V. NEELIMA DEVI

Generative adversarial networks (GANs) provide a way to learn deep representations without extensively annotated training data. They achieve this by deriving backpropagation signals through a competitive process involving a pair of networks. CNN are specialized neural networks responsible for the processing of data with an input shape resembling a 2D matrix like image. They are made use of in image detection or pattern detection and image classification. Images or pictures are basically pixels in a matrix shape, and the trained distance function of the CNN is useful for identifying the image, and/or for classifying the image as per annotated categories . The CNN is a convoluted neural network, where convolution is basically a mathematical operation with input I and argument, and a kernel K producing an output useful for understanding how shapes are modified. "A convolution is a weighted sum of the pixel values of the image, as the window slides across the whole image. Turns out, this convolution process throughout an image with a weight matrix produces another image (of the same size, depending on the convention). Convolving is the process of applying a convolution " . A feature map is drawn with the convolution function. If there is an image x in a given image with pixels in the 2D format, then the array is set out in different color channels of the RGB. A feature detector (also called the kernel) as represented by’ will output the feature map. From the function below, s[t] means feature map, x means input and w means kernel.

nlp database, computer vision, supervised learning, unsupervised learning

"Generative adversiral network using machine learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 6, page no.j312-j324, June-2023, Available :http://www.jetir.org/papers/JETIR2306945.pdf

"Generative adversiral network using machine learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 6, page no. ppj312-j324, June-2023, Available at : http://www.jetir.org/papers/JETIR2306945.pdf