University of Illinois at Chicago
MISHRA-THESIS-2023.pdf (9.94 MB)

Towards Understanding Deep Generative Models for Continuous Data Augmentation

Download (9.94 MB)
posted on 2023-05-01, 00:00 authored by Harsh Mishra
Score-based models, a class of Deep Generative models, are considered state-of-the-art for image generation. The key idea behind such models is to gradually add noise to the data and then use a neural network to recover the original data by reversing the noising process. From a computational perspective, existing score-based models can be efficiently trained only if the forward or the corruption process comprises Gaussian noise, i.e., can be computed in closed form. On this, we first propose the use of continuous data augmenting methods, which provide the alternative to Gaussian noise. We also propose a new framework, named Intermediate Generator Optimization (IGO), that explicitly models such Non-Gaussian forward processes, by using the relationship between the corruption process and the layers in a deep Generative model. The main advantage of our framework is that it can be incorporated into any standard autoencoder pipeline for generative tasks. We provide implementation details to apply our framework on benchmark image generation and point-cloud denoising models, as well as the downstream task of Generative PCA.



Ravi, Sathya N.


Ravi, Sathya N.


Computer Science

Degree Grantor

University of Illinois at Chicago

Degree Level

  • Masters

Degree name

MS, Master of Science

Committee Member

Trivedi, Amit R Vamanan, Balajee Parde, Natalie

Submitted date

May 2023

Thesis type



  • en

Usage metrics


    No categories selected