知识卡片

普通读书笔记卡 · 9254.d · 生成模型:GPT-6

初始质心影响局部收敛结果

理解等级:中级

交替优化可能收敛到不同局部解,初始位置因此重要。尽量分散的初始化可有帮助,但“越远一定越好且一致”不成立;应明确初始化方法并比较多次运行的目标与稳定性。

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

      K-means is a simple unsupervised learning algorithm. It classifies a given data set into K clusters where k is the number of clusters which is predetermined. The algorithm puts the data points into predefined clusters, K which is the input to this algorithm. Next, each data point is assigned to the nearest cluster center, called centroids. Figure 7.1 shows the clusters with the black points

    参考来源

    - 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第07章《无监督学习模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/20_chapter07.xhtml) - 结论依据:摘录中关于初始质心影响局部收敛结果的定义、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:K-means is a simple unsupervised learning algorithm. It classifies a given data set into K clusters where k is the number of clusters which is predetermined. The algorithm puts the data points into predefined clusters, K which is the input to this algorithm. Next, each data point is assigned to the nearest cluster center, called centroids. Figure 7.1 shows the clusters with the black points