Machine Learning (ML) has provided powerful tools for many modern applications, and becomes one of the fundamental approaches to address various issues. As a result, the security of ML methods has become a focus of adversaries to manipulate real-world ML solutions. Generative adversarial networks (GANs) are one serious attacks to many ML solutions. Here, we first summarize the state-of-art GANs, discuss their ideas, and identify their advantages and disadvantages. The goal of this project is to identify research issues in GANs and propose corresponding solutions.
Here is our starting list of readings:
A Beginner's Guide to Generative Adversarial Networks (GANs)
A Gentle Introduction to Generative Adversarial Networks (GANs)
MS Student: Yeon Sang You's notes
Tasks: (1) tutorial presentation of GANs; (2) demo GAN cases with simulations