Generative AI has become ubiquitous, powering tools like ChatGPT and Midjourney, but the true origin of this revolution lies in Generative Adversarial Networks (GANs). In a live session scheduled for May 4, 2026, at 8:00 PM IST, Priya Bhatia will explain the architecture that started it all.
GANs were first introduced by Ian Goodfellow in 2014 and consist of two neural networks—a generator and a discriminator—that compete against each other. The generator creates fake data, while the discriminator tries to distinguish real from fake. This adversarial process drives the generator to produce increasingly realistic outputs, enabling applications like image generation, deepfakes, and data augmentation.
The session promises a practical breakdown of how GANs work, their evolution, and their role in modern AI. Whether you're a beginner or an experienced practitioner, this live discussion will demystify the core concepts behind generative models.