Epistemic Generative Adversarial Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mubashar, Muhammad, Cuzzolin, Fabio
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914408623505408
author Mubashar, Muhammad
Cuzzolin, Fabio
author_facet Mubashar, Muhammad
Cuzzolin, Fabio
contents Generative models, particularly Generative Adversarial Networks (GANs), often suffer from a lack of output diversity, frequently generating similar samples rather than a wide range of variations. This paper introduces a novel generalization of the GAN loss function based on Dempster-Shafer theory of evidence, applied to both the generator and discriminator. Additionally, we propose an architectural enhancement to the generator that enables it to predict a mass function for each image pixel. This modification allows the model to quantify uncertainty in its outputs and leverage this uncertainty to produce more diverse and representative generations. Experimental evidence shows that our approach not only improves generation variability but also provides a principled framework for modeling and interpreting uncertainty in generative processes.
format Preprint
id arxiv_https___arxiv_org_abs_2603_18348
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Epistemic Generative Adversarial Networks
Mubashar, Muhammad
Cuzzolin, Fabio
Machine Learning
Computer Vision and Pattern Recognition
I.2.10
Generative models, particularly Generative Adversarial Networks (GANs), often suffer from a lack of output diversity, frequently generating similar samples rather than a wide range of variations. This paper introduces a novel generalization of the GAN loss function based on Dempster-Shafer theory of evidence, applied to both the generator and discriminator. Additionally, we propose an architectural enhancement to the generator that enables it to predict a mass function for each image pixel. This modification allows the model to quantify uncertainty in its outputs and leverage this uncertainty to produce more diverse and representative generations. Experimental evidence shows that our approach not only improves generation variability but also provides a principled framework for modeling and interpreting uncertainty in generative processes.
title Epistemic Generative Adversarial Networks
topic Machine Learning
Computer Vision and Pattern Recognition
I.2.10
url https://arxiv.org/abs/2603.18348