论文标题
光谱扩散过程
Spectral Diffusion Processes
论文作者
论文摘要
事实证明,基于得分的生成建模(SGM)是在有限维空间上建模密度的一种非常有效的方法。在这项工作中,我们建议将这种方法扩展到在功能空间上学习生成模型。为此,我们代表光谱空间中的功能数据,以将过程的随机部分与时空部分解离。然后,使用有限的尺寸SGM,我们使用尺寸降低技术从其随机组件中取样。我们证明了我们的方法对各种多模式数据集进行建模的有效性。
Score-based generative modelling (SGM) has proven to be a very effective method for modelling densities on finite-dimensional spaces. In this work we propose to extend this methodology to learn generative models over functional spaces. To do so, we represent functional data in spectral space to dissociate the stochastic part of the processes from their space-time part. Using dimensionality reduction techniques we then sample from their stochastic component using finite dimensional SGM. We demonstrate our method's effectiveness for modelling various multimodal datasets.