Generative Adversarial Network Simplified #artificialintelligence #machinelearning #deeplearning

Key Takeaways

Hello friends, I hope you all are doing very well and here in this lecture we are going to simplify the concept of generative adversarial network. It is a deep learning framework where we are having concept of two neural network. One is generator and other one is discriminator. Normally we use GANs for generating highly realistic data. For example, if you want to generate highly realistic image from a given image data set or if you want to generate highly realistic music from the music database, then you can utilize GAN for this. Now, how it works? First thing is that it is called as adversarial because here we are competing two neural networks against each other. One is generator which is going to keep generating the fake data and discriminator which is which is keep going to identify the real data or fake data. And this process will keep going on until and unless our predictive model will unable to identify or unable to distinguish whether the data generated is fake or real. That is what is purpose of GAN. Now to understand this particular topic in detail, please subscribe my YouTube channel Professor Rahul Jain. The link is given into the description section and you can go with the detailed architecture and everything. Thank you so much guys. I hope the simplified version of this lectures are going to be utilized for your effective study and for your comment and suggestions, feel free to utilize the comment section. Thank you so much. Have a very nice day. Jai Hind. Jai Bharat.