• Understanding Diffusion Models: A Unified Perspective

    Diffusion models have shown incredible capabilities as generative models; indeed, they power the current state-of-the-art models on text-conditioned image generation such as Imagen and DALL-E 2. In this tutorial we review, demystify, and unify the understanding of diffusion models across both variational and score-based perspectives. A PDF version of this blog post, for printability, can be found here.

    Continue reading...
  • A Brief Introduction To Information

    Information theory is often glossed over in most introduction to machine learning courses, despite underpinning many useful concepts encountered in the field. A proper understanding of the information theory is quite helpful in providing intuition and perspective behind many of the current ideas in machine learning. My hope is for this post to deliver the basics in an easy-to-digest way.

    Continue reading...
  • Hello World

    Welcome to my website!

    Continue reading...