知识卡片
共享实体因子连接辅助矩阵
协同分解把主张量与辅助矩阵中代表同一实体的潜在因子共享。例如地点同时出现于签到张量和地点—天气矩阵,可通过共享地点表示传递信息。共享前必须对齐实体与尺度,错误对齐会把不同对象混合。
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
only the factor matrices share parameters for gathering knowledge from those feature matrices. Finally, the optimization done by Stochastic Gradient Descent (SGD) [9] considers non-zero elements to fetch the minimum values and output generated, which is nothing but the four-factor matrices.
参考来源
- 位置:《Artificial Intelligence and Data Science in Recommendation System: Current Trends, Technologies, and Applications》第11章《张量分解的兴趣点推荐》(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-C11.html)
- 结论依据:摘录中关于共享实体因子连接辅助矩阵的定义、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:only the factor matrices share parameters for gathering knowledge from those feature matrices. Finally, the optimization done by Stochastic Gradient Descent (SGD) [9] considers non-zero elements to fetch the minimum values and output generated, which is nothing but the four-factor matrices.