The perceptron model y = f ( x ) = sign ( w T x + b
Question:
The perceptron model y=f(x)=sign(wTx+b)y=f(x)=sign(wTx+b) can be used to learn a binary classifier from training data.
a. Assume there are two training samples. The positive one is x1=(2,1)Tx1=(2,1)T; the negative one is x2=(1,0)Tx2=(1,0)T. The learning rate η=1η=1. Starting from w=(1,1)Tw=(1,1)T and b=0b=0, solve the parameters of the classifier.
b. Assume there are four training samples. The positive samples are x1=(1,1)Tx1=(1,1)T and x2=x2= (0,0)T(0,0)T; the negative samples are x3=(1,0)Tx3=(1,0)T and x4=(0,1)Tx4=(0,1)T. Can we classify all training samples correctly using the perceptron model? Why?
Fantastic news! We've Found the answer you've been seeking!
Step by Step Answer:
Related Book For
Data Mining Concepts And Techniques
ISBN: 9780128117613
4th Edition
Authors: Jiawei Han, Jian Pei, Hanghang Tong
Question Posted: