Generalization vs. Hallucination
Abstract
With fast developments in computational power and algorithms, deep learning has made breakthroughs and been applied in many fields. However, generalization remains to be a critical challenge, and the limited generalization capability severely constrains its practical applications. Hallucination issue is another unresolved conundrum haunting deep learning and large models. By leveraging a physical model of imaging through scattering media, we studied the lack of generalization to system response functions in deep learning, identified its cause, and proposed a universal solution. The research also elucidates the creation process of a hallucination in image prediction and reveals its cause, and the common relationship between generalization and hallucination is discovered and clarified. Generally speaking, it enhances the interpretability of deep learning from a physics-based perspective, and builds a universal physical framework for deep learning in various fields. It may pave a way for direct interaction between deep learning and the real world, facilitating the transition of deep learning from a demo model to a practical tool in diverse applications.
Cite
@article{arxiv.2411.02893,
title = {Generalization vs. Hallucination},
author = {Xuyu Zhang and Haofan Huang and Dawei Zhang and Songlin Zhuang and Shensheng Han and Puxiang Lai and Honglin Liu},
journal= {arXiv preprint arXiv:2411.02893},
year = {2024}
}