知识卡片

普通读书笔记卡 · 6488.q · 生成模型:GPT-6

多层注意力区分情境来源

理解等级:中级

多层情境注意力可以在不同来源及来源内部的因素之间分配学习权重,让下一地点预测随上下文变化。权重是模型计算的一部分,不自动解释因果,也不能单凭权重大认定某因素现实中最重要。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      After embedding all the factors into latent vectors, these are fed into the encoder to learn about different contexts, then into the decoder to predict the next POI to be visited using encoder hidden states. For the final prediction, multi-level context attention is used to get better attention on different factors and hierarchical attention with two levels: macro context attention and micro context attention.

    参考来源

    - 位置:《Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications》第09章《兴趣点推荐的影响参数》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications/_epub-src/OEBPS/Text/9789815136746-C9.html) - 结论依据:摘录中关于多层注意力区分情境来源的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:After embedding all the factors into latent vectors, these are fed into the encoder to learn about different contexts, then into the decoder to predict the next POI to be visited using encoder hidden states. For the final prediction, multi-level context attention is used to get better attention on different factors and hierarchical attention with two levels: macro context attention and micro context attention.