知识卡片
CP分解累加秩一项
CP分解用若干秩一张量相加近似原张量,把多轴关系表示为对应潜在因子的共同作用。所选项数影响表达与开销;近似项数由建模决定,不意味着已经证明原数据的精确最小秩。
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
There are different types of tensor decomposition available, mainly based on the two most important decomposition techniques: canonical polyadic decomposition (CPD) and the Tucker decomposition (TD). Both are outer product decomposition [18] but have different structural properties [20]. The CPD is a rank-decomposition [17, 18] which can be defined as rank-one tensors with the sum of a finite number [18].
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:摘录中关于正则多项分解累加秩一项的定义、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:There are different types of tensor decomposition available, mainly based on the two most important decomposition techniques: canonical polyadic decomposition (CPD) and the Tucker decomposition (TD). Both are outer product decomposition [18] but have different structural properties [20]. The CPD is a rank-decomposition [17, 18] which can be defined as rank-one tensors with the sum of a finite number [18].