Machine Learning-Perception machine

Summary:

1. Introduction

2. Model

3. Strategy

4. Algorithms

4.1 Original Questions

4.2 Duality problem

Content:

1. Introduction

The Perceptron is a linear classification model of two classification, and the output is +1,-1. The discrete hyper-plane of the perceptual machine corresponding to the input space belongs to the discriminant model. Perceptron is the basis of neural network and support vector machine.

2. Model

Or, where

3. Strategy

　　　　

4. Algorithms

4.1 Original Questions

Input: Linear sub-dataset, learning rate A (0<a<=1)

(1) Select the initial value w0,b0=0

(2) Selecting data in the training set (Xi,yi)

(3) If the W:=w+ayixi,b:=b+ayi

(4) Loop Execution (2) ~ (3) until no wrong classification point

4.2 Duality problem

　　　　　　　

　　　

　

Machine Learning-Perception machine