机器学习相关数据基础

全是转载

 

1.最小二乘法：https://www.zhihu.com/question/37031188

2.矩阵求导公式：https://blog.csdn.net/a493823882/article/details/81324037?utm_medium=distribute.pc_relevant_t0.none-task-blog-BlogCommendFromMachineLearnPai2-1.nonecase&depth_1-utm_source=distribute.pc_relevant_t0.none-task-blog-BlogCommendFromMachineLearnPai2-1.nonecase

3.先验概率与后验概率：https://www.cnblogs.com/yemanxiaozu/p/7680761.html

4.极大似然估计：https://blog.csdn.net/weixin_42159940/article/details/91077943

5.梯度下降：https://zhuanlan.zhihu.com/p/43452377  https://www.jianshu.com/p/c7e642877b0e

6.RNN的time_step和batch_size的理解：https://blog.csdn.net/lnaruto1234/article/details/99672601

7.为什么权重总是初始化为正态分布且标准差为0.01 https://blog.csdn.net/Harpoon_fly/article/details/84844621

8.泰勒展开式 https://www.zhihu.com/question/25627482

9.信息熵为什么用对数 https://blog.csdn.net/taoqick/article/details/72852255

10.为什么要有激活函数 https://www.zhihu.com/question/22334626

11.softmax https://blog.csdn.net/bitcarmanlee/article/details/82320853