<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>图神经网络 on C.CUI's Log</title><link>https://cuicaihao.github.io/zh/tags/%E5%9B%BE%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C/</link><description>Recent content in 图神经网络 on C.CUI's Log</description><generator>Hugo</generator><language>zh-Hans</language><lastBuildDate>Fri, 30 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://cuicaihao.github.io/zh/tags/%E5%9B%BE%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C/index.xml" rel="self" type="application/rss+xml"/><item><title>用图神经网络做零售需求预测：数学原理、受控实验与生产落地边界</title><link>https://cuicaihao.github.io/zh/posts/2026-10-24-fsm-stochastic-gnn-retail-forecast/</link><pubDate>Sat, 24 Oct 2026 07:00:00 +1100</pubDate><guid>https://cuicaihao.github.io/zh/posts/2026-10-24-fsm-stochastic-gnn-retail-forecast/</guid><description>受控实验显示，修复销量截断和加入一跳邻域信息往往比直接部署 GNN 更重要。只有跨店耦合真实、输入及时、多跳结构仍有稳定增量时，图模型才值得承担额外复杂度。</description></item></channel></rss>