通过温度调节提升扩散采样器探索能力，改善采样质量。
Continuously Tempered Diffusion Samplers
- 引入多温度分布降低能量壁垒，增强采样探索性。
- 在多个数据集上实现更高采样精度，优于传统方法。
- 适合需要高质量生成样本的研究者使用。
基于退火的神经采样器通过训练神经网络将一系列从源到目标密度的插值分布进行传输，以实现对未归一化分布的高效采样。此类采样器训练中的关键设计在于损失评估位置的提议分布。以往工作通过结合部分学习的传输与退火朗之万动力学来生成提议分布，但退火路径中存在的孤立模式等病态特性导致提议分布探索不足，从而影响训练后性能。为解决此问题，本文提出连续温度调节扩散采样器（Continuously Tempered Diffusion Samplers），借鉴分子动力学中的探索技术，在不同温度下引入分布族，以在高温时降低能量壁垒并驱动探索，而在目标低温下实现高效采样。实验验证了扩展探索带来的采样性能提升。代码已开源：https://github.com/eje24/ctds。
原文摘要 · Abstract (English)
Annealing-based neural samplers seek to amortize sampling from unnormalized distributions by training neural networks to transport a family of densities interpolating from source to target. A crucial design choice in the training phase of such samplers is the proposal distribution by which locations are generated at which to evaluate the loss. Previous work has obtained such a proposal distribution by combining a partially learned transport with annealed Langevin dynamics. However, isolated modes and other pathological properties of the annealing path imply that such proposals achieve insufficient exploration and thereby lower performance post training. To remedy this, we propose continuously tempered diffusion samplers, which leverage exploration techniques developed in the context of molecular dynamics to improve proposal distributions. Specifically, a family of distributions across different temperatures is introduced to lower energy barriers at higher temperatures and drive exploration at the lower temperature of interest. We empirically validate improved sampler performance driven by extended exploration. Code is available at https://github.com/eje24/ctds.
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。