《机器学习基石》台大林轩田_学习笔记02_Learning to Answer Yes/No

Lecture 2: Learning to Answer Yes/No

大纲
1. Perceptron Hypothesis Set
2. Perceptron Learning Algorithm(PLA)
3. Guarantee of PLA
4. Non-Separable Data

1. Perceptron Hypothesis Set
perceptrons 感知机<=>linear(binary) classifiers
For x=(x1,x2,...,xn)
Y:+1,1
h(x)=sign((ni=1wixi)θ)=sign(ni=0wixi)=sign(wTx)

2. Perceptron Learning Algorithm(PLA)
知错能改算法
《机器学习基石》台大林轩田_学习笔记02_Learning to Answer Yes/No
3. Guarantee of PLA

4. Non-Separable Data