📋 Overview
Research intern at UNIST Embedded AI Lab, where I reviewed DDPM papers, built a toy diffusion project on MNIST, and explored sampling optimization directions.
🛠️ Technologies
⚡ Key Projects & Activities
- DDPM Paper Review - Reviewed the Diffusion Model (DDPM) paper and organized key implementation ideas
- MNIST Toy Training - Trained a diffusion model on the MNIST dataset as a hands-on toy project
- DDIM + CFG Extension - Applied DDIM sampling and classifier-free guidance to generate target digit images from user input
- Timestep Optimization Study - Explored timestep sampling optimization research; no significant gains, but strong first image-generation project experience