My GRADflix project is a playful 8-bit pixel-style video game, designed to present my MMath research on visualizing adversarial machine learning in an accessible and creative way. I developed the game using a game engine for the logic and interactivity, and hand-drew all the artwork from scratch using Procreate on my iPad. The game is set to autoplay, and I recorded a playthrough to demonstrate how I can combine my programming and design skills to make complex AI concepts more understandable to a wider audience. The goal of this project is to highlight my research and emphasize the value of visualizations in explaining AI safety, as well as making technical topics more engaging for non-technical audience to grasp.
- Selected as a finalist in the 2023 University of Waterloo GRADflix competition.
- UWaterloo News: Meet the GRADflix finalist who combined art and programming .
- A 8-bit pixel animation combining art and programming.
- Custom game engine scripting and programming for seamless animations and interactions.
- Interactive video game elements used to explain adversarial machine learning and XAI visualization research.
- Created a fully functional game, set to autoplay and recorded as a screencast.
Meet the GRADflix finalist who combined art and programming.
- Pixel Art
- Game Development
- Game Design
- AI Interpretability
- Procreate
- Game Engine Programming
- Adversarial Machine Learning
- Visualization
- Pixel Art
- Game Development
- AI Interpretability