吴恩达《机器学习》笔记01 - 机器学习的动机与应用

机器学习 定义

Arthur Samuel (1959): Field of study that gives computers the ability to learn without being explicitly programed.

Tom Mitchell (1998) Well-posed Learning Problem: A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E.

课堂中会谈到的四个方面

1. 监督学习 Supervised learning

• 回归问题 Regression problem
  需要预测的变量是连续的
  例如：根据房屋面积预测其价格
  

• 分类问题 Classification problem
  需要预测的变量是离散的 (SVM)
  例如：根据肿瘤大小及年龄预测肿瘤是良性还是恶性
  

2. 学习理论 Learning theory

3. 非监督学习 Unsupervised learning

用于分类，如计算机视觉中的应用
ICA 算法: [W,s,v]=svd((repmat(sum(x.*x,1),size(x,1),1).*x)*x’);

4. 强化学习 Reinforcement learning