知识卡片
可观测性限制情境推断
推荐时某些情境可能已知,某些缺失,或全部需要推断。方法必须说明部署时实际能看到什么;离线拥有完整情境而线上缺失,会让评估条件不成立。缺失值也不宜伪装成确定的普通类别。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
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
Partially observable: When contextual dimensions suffer from “partial observability”, which may lead to incomplete contextual information. For instance, if only the Temporal dimension is available for a given set of users, all other contextual dimensions are available for another group [52].
参考来源
- 位置:《Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications》第04章《多领域情境推荐的张量分解》(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-C4.html)
- 结论依据:摘录中关于可观测性限制情境推断的定义、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:Partially observable: When contextual dimensions suffer from “partial observability”, which may lead to incomplete contextual information. For instance, if only the Temporal dimension is available for a given set of users, all other contextual dimensions are available for another group [52].