Diffusion models struggle to produce samples that respect constraints, a common requirement in scientific applications. Recent approaches have introduced regularization terms in the loss or guidance methods during sampling to enforce such constraints, but they bias the generative model away from the true data distribution. This is a problem, especially when the constraint is misspecified, a common issue when formulating constraints on scientific data. In this paper, instead of changing the loss or the sampling loop, we integrate a guidance-inspired adjustment into the denoiser itself, giving it a soft inductive bias towards constraint-compliant samples. We show that these softly constrained denoisers exploit constraint knowledge to improve compliance over standard denoisers, and maintain enough flexibility to deviate from it when there is misspecification with observed data.
翻译:扩散模型在生成满足约束条件的样本方面存在困难,而这在科学应用中是一个常见需求。近期方法通过在损失函数中引入正则化项或在采样过程中采用引导方法来强制执行此类约束,但这些方法会使生成模型偏离真实数据分布。当约束条件设定错误时(这在科学数据约束建模中是常见问题),这尤其成为一个难题。本文提出一种新方法:我们不改变损失函数或采样循环,而是将引导式调整直接集成到去噪器内部,使其对符合约束的样本产生软性归纳偏置。研究表明,这些软约束去噪器能够利用约束知识提升对标准去噪器的约束遵循度,同时在约束条件与观测数据存在失配时保持足够的灵活性以偏离约束。