Blogs - Duy-Tan Nguyen

Principles of Deep Learning

An overview of fundamental principles in deep learning, including backpropagation, gradient descent, and regularization techniques. (This blog is still a work in progress.)

Overview of Diffusion Models

This article delves into Diffusion Models—a powerful class of generative models that has revolutionized the field of generative AI. We will explore their core principles, theoretical foundations, and practical applications, thereby highlighting their unique capabilities and the reasons behind their widespread adoption within the AI community.

Advanced Diffusion Models

This article delves into advanced concepts regarding Diffusion Models—a class of generative models that has revolutionized the field of generative AI. We will explore the mathematical foundations of these models. The article also covers in-depth topics such as the flow matching framework, optimal transport, Schrödinger bridges, sampling techniques, and sophisticated solvers.

Diffusion Models: Probability Flow ODE and Probability Path

(This blog is still a work in progress.)

From Variational Bayes to Diffusion Models: The Journey of Generative Mathematics

This article traces a mathematical thread connecting the fundamental problem of Bayesian statistics—the intractability of the posterior distribution—to the denoising objective that lies at the heart of modern Diffusion Models.