Epistemic Generative Adversarial Networks

2026-03-18

AI Summary

This work addresses the issue of mode collapse in Generative Adversarial Networks (GANs), which often leads to insufficient diversity in generated samples. For the first time, the Dempster–Shafer evidence theory is integrated into the GAN framework by reformulating the loss function and embedding a pixel-level mass function prediction mechanism within the generator. This approach explicitly models output uncertainty, thereby not only enhancing the diversity of generated samples but also providing an interpretable quantification of uncertainty throughout the generation process. Consequently, the proposed method strengthens the model’s representational capacity and expressive power while maintaining fidelity to the underlying data distribution.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

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.

Problem

• Generative Adversarial Networks
• output diversity
• sample variability
• uncertainty quantification
• generative models

Innovation

• Epistemic Uncertainty
• Dempster-Shafer Theory
• Generative Adversarial Networks
• Mass Function
• Uncertainty Quantification