arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2401.03140cs.LGcs.CY
收藏

通过切换机制的扩散模型公平采样

Fair Sampling in Diffusion Models through Switching Mechanism

Seoul National University(首尔国立大学);Ulsan National Institute of Science and Technology (UNIST)(蔚山国立科学技术院(UNIST))

Yujin Choi, Jinseong Park, Hoki Kim, Jaewook Lee, Saerom Park

AI总结:

本文提出一种无需额外训练的扩散模型属性切换采样机制,在不依赖分类器的情况下混淆敏感属性,实现公平数据生成并保持数据效用。

摘要:

扩散模型通过很好地逼近底层概率分布,在生成任务中展现了其有效性。然而,众所周知,扩散模型在公平性方面会遭受来自训练数据的固有偏见的放大。虽然扩散模型的采样过程可以通过条件引导来控制,但先前的工作试图寻找经验性引导以实现定量公平。为了解决这一局限性,我们提出了一种面向公平的采样方法,称为扩散模型的属性切换机制。在无需额外训练的情况下,所提出的采样可以在不依赖分类器的情况下混淆生成数据中的敏感属性。我们在两个关键方面数学证明并通过实验证明了所提方法的有效性:(i)公平数据的生成和(ii)生成数据效用的保持。

英文摘要:

Diffusion models have shown their effectiveness in generation tasks by well-approximating the underlying probability distribution. However, diffusion models are known to suffer from an amplified inherent bias from the training data in terms of fairness. While the sampling process of diffusion models can be controlled by conditional guidance, previous works have attempted to find empirical guidance to achieve quantitative fairness. To address this limitation, we propose a fairness-aware sampling method called \textit{attribute switching} mechanism for diffusion models. Without additional training, the proposed sampling can obfuscate sensitive attributes in generated data without relying on classifiers. We mathematically prove and experimentally demonstrate the effectiveness of the proposed method on two key aspects: (i) the generation of fair data and (ii) the preservation of the utility of the generated data.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑