Scribd
Upload
8 views36 pages

SVM_linear

The document discusses support vector machines (SVM) with a focus on non-linear kernel functions and their architecture. It includes examples of linearly inseparable data, the concept of discriminating hyperplanes, and multi-class SVM strategies like one-against-one classification. Additionally, it highlights the representation of data in feature space and the use of support vectors in finding parameters.

Uploaded by

Rknif 781
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
8 views36 pages

SVM_linear

The document discusses support vector machines (SVM) with a focus on non-linear kernel functions and their architecture. It includes examples of linearly inseparable data, the concept of discriminating hyperplanes, and multi-class SVM strategies like one-against-one classification. Additionally, it highlights the representation of data in feature space and the use of support vectors in finding parameters.

Uploaded by

Rknif 781
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 36

Support vector machines:

Non-Linear, Kernel Functions


Architecture
SVM Architecture
SVM-Example

Nonlinear function
SVM – Linearly Inseparable data - Example
Nonlinearly separable sample data points -
Example
Example-2
Vectors
Augmented Support Vector
Find Parameters by Substituting Support
vectors
Discriminating Hyper plane
Separating
Non-Linear SVM – Example (cont’d)
Example (cont’d)
Data represented in feature space -
Example (cont’d)
Data represented in feature space -
Example (cont’d)
• The two support vectors (in feature space) are marked as
yellow circles.
Example (cont’d)
Example (cont’d)
Hyperplane - Example (cont’d)
• The discriminating hyperplane corresponding to the values α1
= -7 and α2 = 4
Multi-Class SVM

• One-against-one
– Distinguish one class from another one.
– Train one classifier per pair of classes, which leads
to K(K-1)/2 classifiers for K classes.
– Each classifier is trained on a subset of the data
and produces its own decision boundary
– Predictions are made using a simple voting
strategy
Examples of Kernel Functions

You might also like