eathotpot的博客

机器学习笔记_introduction——Oct30 2020 一、任务类型 regression 回归问题 输出为数字 binary classification 二元分类yes or no(pos or neg) multi-class classification 多元分类到class1、class2…classN generation 生成 二、学习方式 supervise learning/监督学习 and reinforcement learning/强化学习 ...

1.Model 1.1 Linear model y=b+∑(wixi)y=b+\sum(w_{i}x_{i})y=b+∑(wi​xi​) w3w_{3}w3​:wight bbb:bias xix_{i}xi​:input 1.2 Cost function L(f)=L(w,b)=∑n=1N(y^n−(b+∑w∗xn))2L(f)=L(w,b)=\sum_{n=1}^N(\hat y^n-(b+\sum w*x^n))^2L(f)=L(w,b)=n=1∑N​(y^​n−(b+∑w∗xn))2 (真实值−