让扩散模型一步生成符合物理规律的样本
Consistency Training with Physical Constraints
- 用一致性训练学习噪声到数据的映射
- 单步生成且满足物理约束，实验验证有效
- 适合需要快速求解偏微分方程的科研与工程场景
我们提出一种融合物理约束的一致性训练（Consistency Training, CT）方法，可加速带有物理约束的扩散模型采样。该方法采用两阶段策略：（1）通过一致性训练学习噪声到数据的映射；（2）将物理约束作为正则项融入训练过程。在简化示例上的实验表明，该方法可在单步内生成符合给定约束条件的样本。此方法有望高效利用深度生成建模求解偏微分方程（PDEs）。
原文摘要 · Abstract (English)
We propose a physics-aware Consistency Training (CT) method that accelerates sampling in Diffusion Models with physical constraints. Our approach leverages a two-stage strategy: (1) learning the noise-to-data mapping via CT, and (2) incorporating physics constraints as a regularizer. Experiments on toy examples show that our method generates samples in a single step while adhering to the imposed constraints. This approach has the potential to efficiently solve partial differential equations (PDEs) using deep generative modeling.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。