用散射表示直接做仿真推断，无需额外模拟或求导。
Simulation-based inference with scattering representations: scattering is all you need
- 用散射变换直接提取图像特征，避免压缩损失
- 在宇宙学案例中比传统统计量提取更多信息
- 无需额外训练数据，对分布偏移有强鲁棒性
我们展示了在基于仿真的推断（SBI）中，直接使用散射表示进行图像级（场级）推断的成功应用，以宇宙学案例为例。散射表示为后续学习任务提供了高效的表征空间，尽管高维压缩空间带来挑战。通过结合空间平均与更强大的密度估计器，我们克服了这些困难。相比其他方法，该方案无需额外仿真用于训练或计算导数，具有可解释性且对协变量偏移具有鲁棒性。如预期，仅使用散射的方法提取的信息量超过传统二阶摘要统计量。
原文摘要 · Abstract (English)
We demonstrate the successful use of scattering representations without further compression for simulation-based inference (SBI) with images (i.e. field-level), illustrated with a cosmological case study. Scattering representations provide a highly effective representational space for subsequent learning tasks, although the higher dimensional compressed space introduces challenges. We overcome these through spatial averaging, coupled with more expressive density estimators. Compared to alternative methods, such an approach does not require additional simulations for either training or computing derivatives, is interpretable, and resilient to covariate shift. As expected, we show that a scattering only approach extracts more information than traditional second order summary statistics.
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。