Gurprit Singh

Abstract

Generative artificial intelligence (AI) has made unprecedented advances in vision language models over the past two years. These advances are largely due to diffusion-based generative models, which are very stable and simple to train. These diffusion models are tasked to learn the underlying unknown distribution of the training data samples. During the generative process, new samples (images) are generated from this unknown high-dimensional distribution. Markov Chain Monte Carlo (MCMC) methods are particularly effective in drawing samples from complex, high-dimensional distributions. This makes MCMC methods an integral component for both the training and sampling phases of these models, ensuring accurate sample generation.
Gradient-based optimization is at the core of modern generative models. The update step during the optimization forms a Markov chain where the new update depends only on the current state. This allows exploration of the parameter space in a memoryless manner, thus combining the benefits of gradient-based optimization and MCMC sampling. MCMC methods have shown an equally important role in physically based rendering where complex light paths are otherwise quite challenging to sample from simple importance sampling techniques. A lot of research is dedicated towards bringing physical realism to samples (images) generated from diffusion-based generative models in a data-driven manner, however, a unified framework connecting these techniques is still missing. In this course, we take the first steps toward understanding each of these components and exploring how MCMC could potentially serve as a bridge, linking these closely related areas of research. Our tutorial aims to provide necessary theoretical and practical tools to guide students, researchers and practitioners towards the common goal of generative physically based rendering.

MCMC: Bridging Rendering, Optimization and Generative AI

Abstract

Material

Copyright Disclaimer