Synthetic Data Generation with GANs - Quant Researcher Interview Question

Problem Description

Generative Adversarial Networks (GANs) are essential tools in quantitative finance for creating synthetic market data, enabling the stress testing of trading strategies and the training of reinforcement learning agents on simulated scenarios. By orchestrating a zero-sum game between a generator and a discriminator, these models learn to approximate complex asset return distributions. Task Implement a simultaneous Gradient Descent update step for a simple 1D GAN consisting of a linear generator