Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2411.02893 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910685583114240 |
|---|---|
| author | Zhang, Xuyu Huang, Haofan Zhang, Dawei Zhuang, Songlin Han, Shensheng Lai, Puxiang Liu, Honglin |
| author_facet | Zhang, Xuyu Huang, Haofan Zhang, Dawei Zhuang, Songlin Han, Shensheng Lai, Puxiang Liu, Honglin |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_02893 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Generalization vs. Hallucination Zhang, Xuyu Huang, Haofan Zhang, Dawei Zhuang, Songlin Han, Shensheng Lai, Puxiang Liu, Honglin Optics 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. |
| title | Generalization vs. Hallucination |
| topic | Optics |
| url | https://arxiv.org/abs/2411.02893 |