发表机构
Argonne National Laboratory; SLAC National Accelerator Laboratory; Lawrence Berkeley National Laboratory（阿贡国家实验室; SLAC国家加速器实验室; 劳伦斯伯克利国家实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对叠层成像神经网络分布外泛化时的缩放不一致问题，提出分解策略与合成物体采样策略，在5个实验数据集上将傅里叶误差较PtychoPINN-torch基线降低最多5倍，提升了其实际应用可行性。
AI 中文摘要
叠层成像神经网络在分布外泛化时会出现缩放不一致的问题，限制了其实际应用的可行性。我们采用一种分解策略来解决这种缩放不匹配问题，该策略将学习到的物体纹理与测量缩放解耦，使单个训练好的网络能够在不同照明条件下生成与测量一致的重建结果。这需要用实部和虚部单位来预测学习到的物体，而非采用常规的振幅和相位表示。我们还引入了一种合成物体采样策略，该策略可最小化合成训练数据与实验目标之间的相位分布不匹配。这些改进在涵盖多个光束线和设施的5个实验数据集上，相较于之前的PtychoPINN-torch基线，使傅里叶误差降低了多达5倍。
英文摘要
Ptychography neural networks suffer from scaling inconsistencies when generalizing out of distribution, limiting their real world viability. We address this scaling mismatch using a factorization strategy which decouples the learned object texture from measurement scaling, enabling a single trained network to produce measurement-consistent reconstructions across varying illumination conditions. This requires predicting the learned object in real and imaginary units instead of the canonical amplitude and phase representation. We additionally introduce a synthetic object sampling strategy that minimizes phase distribution mismatch between synthetic training data and experimental targets. These improvements yield up to a 5x reduction in Fourier error over the previous PtychoPINN-torch baseline across 5 experimental datasets spanning multiple beamlines and facilities.
Comments~30 pages, 4 main figures + 2 SI figures. Submitting to scientific journal
