Deeper Diffusion Models Amplify Bias
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909807869427712 |
|---|---|
| author | Hakemi, Shahin Akhtar, Naveed Hassan, Ghulam Mubashar Mian, Ajmal |
| author_facet | Hakemi, Shahin Akhtar, Naveed Hassan, Ghulam Mubashar Mian, Ajmal |
| contents | Despite the remarkable performance of generative Diffusion Models (DMs), their internal working is still not well understood, which is potentially problematic. This paper focuses on exploring the important notion of bias-variance tradeoff in diffusion models. Providing a systematic foundation for this exploration, it establishes that at one extreme, the diffusion models may amplify the inherent bias in the training data, and on the other, they may compromise the presumed privacy of the training samples. Our exploration aligns with the memorization-generalization understanding of the generative models, but it also expands further along this spectrum beyond "generalization", revealing the risk of bias amplification in deeper models. Our claims are validated both theoretically and empirically. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_17560 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Deeper Diffusion Models Amplify Bias Hakemi, Shahin Akhtar, Naveed Hassan, Ghulam Mubashar Mian, Ajmal Computer Vision and Pattern Recognition Despite the remarkable performance of generative Diffusion Models (DMs), their internal working is still not well understood, which is potentially problematic. This paper focuses on exploring the important notion of bias-variance tradeoff in diffusion models. Providing a systematic foundation for this exploration, it establishes that at one extreme, the diffusion models may amplify the inherent bias in the training data, and on the other, they may compromise the presumed privacy of the training samples. Our exploration aligns with the memorization-generalization understanding of the generative models, but it also expands further along this spectrum beyond "generalization", revealing the risk of bias amplification in deeper models. Our claims are validated both theoretically and empirically. |
| title | Deeper Diffusion Models Amplify Bias |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.17560 |