Scribd
Upload
3 views36 pages

Data An-6

Kernel machines are algorithms for pattern analysis that utilize the kernel trick to operate in high-dimensional feature spaces without explicit computation. Various kernel functions, such as linear, polynomial, Gaussian RBF, and Laplace kernels, are discussed, each with unique properties and applications. The choice of kernel is critical for model success, particularly in non-linear data scenarios, and kernel methods remain significant in machine learning.

Uploaded by

chaaru latha
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
3 views36 pages

Data An-6

Kernel machines are algorithms for pattern analysis that utilize the kernel trick to operate in high-dimensional feature spaces without explicit computation. Various kernel functions, such as linear, polynomial, Gaussian RBF, and Laplace kernels, are discussed, each with unique properties and applications. The choice of kernel is critical for model success, particularly in non-linear data scenarios, and kernel methods remain significant in machine learning.

Uploaded by

chaaru latha
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

KERNEL MACHINE

FROM SVM TO KM
Kernel Machines

Kernel Machines

• Kernel machines are a class of algorithms used for


pattern analysis.

• They are based on the concept of the kernel trick, which


allows algorithms to operate in a high-dimensional
feature space without explicitly computing the coordinates
of the data in that space.
Mathematical Representation:
𝐾(𝑥,𝑦)=(𝜙𝑥)⋅𝜙
(𝑦)
• Linear Kernel

• Polynomial Kernel:

Where x and y are the input feature vectors, c is a


constant term, and d is the degree of the polynomial,
K(x, y) = (x. y + c)d
Gaussian Radial Basis Function (RBF) kernel

It measures the similarity between two points in a high-dimensional space.


The formula for the Gaussian RBF kernel is:
• Gaussian (RBF) Kernel radial basis function

Where x and y are the input feature vectors, gamma is a parameter that controls the width of
the Gaussian function, and
||x - y||^2 is the squared Euclidean distance between the input vectors

• Laplace Kernel

Where x and y are the input feature vectors, gamma is a parameter that controls the width of
the
Laplacian function, and ||x - y|| is the L1 norm or Manhattan distance between the input
vectors.
Consider an example to find the similarity between two vectors – ‘x’ and ‘y’, using
Cosine Similarity. The ‘x’ vector has values, x = { 3, 2, 0, 5 } The ‘y’ vector has values,
y = { 1, 0, 0, 0 } The formula for calculating the cosine similarity is :
Cosine_Sim (x, y) = x . y / ||x|| * ||y||
• Characteristics of Kernel Function
• Symmetry: A kernel function is symmetric, meaning that it
produces the same value regardless of the order in which the
inputs are given.
• Reproducing property: it can be used to reconstruct the input
data in the feature space.
• Smoothness: a smooth transformation of the input data into the
feature space.
• Complexity: more complex kernel functions may lead to over
fitting and reduced generalization performance
• Characteristics of Kernel Function
• Symmetry: A kernel function is symmetric, meaning that it
produces the same value regardless of the order in which the
inputs are given.
• Reproducing property: it can be used to reconstruct the input
data in the feature space.
• Smoothness: a smooth transformation of the input data into the
feature space.
• Complexity: more complex kernel functions may lead to over
fitting and reduced generalization performance
Conclusion
- Kernel machines extend linear algorithms to work on non-linear data.
- SVM is a powerful tool when combined with kernel functions.
- Final Thoughts:
- Choosing the right kernel is crucial for the success of the model.
- Kernel methods continue to be relevant in various domains of machine learning.

---

You might also like