Complex data are often represented as a graph, which in turn can often be viewed as a realisation of a random graph, such as an inhomogeneous random graph model (IRG). For general fast goodness-of-fit tests in high dimensions, kernelised Stein discrepancy (KSD) tests are a powerful tool. Here, we develop a KSD-type test for IRG models that can be carried out with a single observation of the network. The test applies to a network of any size, but is particularly interesting for small networks for which asymptotic tests are not warranted. We also provide theoretical guarantees.

翻译：复杂数据常以图的形式呈现，而这类图往往可视为随机图（例如非齐次随机图模型IRG）的实现。针对高维场景下通用的快速拟合优度检验，核化斯坦因差异（KSD）检验是一种强大工具。本文我们针对IRG模型开发了一种仅需网络单次观测即可实施的KSD型检验方法。该检验适用于任意规模的网络，尤其对渐进检验不适用的小型网络具有重要价值。同时，我们提供了理论保证。