In machine learning, a kernel is a function that measures the similarity between two data points in a given feature space. Kernels are used in Support Vector Machines (SVMs) to help perform certain calculations faster.  Here are some types of kernels: Laplacian kernel: Also known as the Laplace kernel or the exponential kernel, this is a non-parametric kernel that can be used to measure the similarity or distance between two input feature vectors. The Laplacian kernel is from the family of RBF kernel and it can be used in noiseless data. Polynomial kernel: This kernel function is commonly... Show more

In machine learning, a kernel is a function that measures the similarity between two data points in a given feature space. Kernels are used in Support Vector Machines (SVMs) to help perform certain calculations faster. 

Here are some types of kernels:

Laplacian kernel: Also known as the Laplace kernel or the exponential kernel, this is a non-parametric kernel that can be used to measure the similarity or distance between two input feature vectors. The Laplacian kernel is from the family of RBF kernel and it can be used in noiseless data.
Polynomial kernel: This kernel function is commonly used with SVMs and other kernelized models. It represents the similarity of vectors (training samples) in a feature space over polynomials of the original variables, allowing learning of non-linear models.
Gaussian kernel: This is a very popular kernel function used in many machine learning algorithms, especially in SVMs. It is more often used than polynomial kernels when learning from nonlinear datasets.
Radial basis function kernel: This kernel function finds a non-linear classifier or regression line. Kernel functions transform n-dimensional input to m-dimensional input, where m is much higher than n. 

Kernel methods are used in SVM to transform nonlinearity present in the dataset to reduce calculation tasks and introduce linearity. 

The kernel trick is a machine learning technique that allows linear models to be applied to nonlinear problems. It does this by transforming data into a higher-dimensional feature space. 

The kernel trick is often used in the support vector machine (SVM) model. The SVM model is a supervised machine learning model that's used for classification and regression. It learns how to separate different groups by forming decision boundaries. 
The kernel trick is based on the observation that the gradient is a linear combination of the input samples when using gradient descent with any standard loss function.

Show less

Machine Learning 101 Practice Test: Kernels And Kernel Trick

Time left 00:00

20 Questions

1. A Hilbert space is a vector space with an inner product, which is also complete.

FalseDon't knowTrueMaybe

2. The Gaussian kernel is also called the RBF kernel, for Radial Basis Functions.

FalseDon't knowTrueMaybe

3. Spectrum Kernel count the number of substrings in common.

FalseTrueDon't knowMaybe

4. Which of the following statements is not true about kernel trick?

It allows one to incorporate prior knowledge of the problem domainThe training data only enter the algorithm through their entries in the kernel matrixThe training data only enter the algorithm through their individual attributesThe number of operations required is not necessarily proportional to the number of features

5. Which of the following statements is not true about kernel?

The kernel function does the mapping of the domain set into a space where the similarities are realised as outer productsThe kernel function K specify the similarity between instancesKernel is used to describe inner products in the feature spaceThe kernel function K specify the embedding as mapping the domain set into a space

6. Which of the following statements is not true about choosing the right kernel?

Linear kernel allows picking out linesA polynomial kernel allows to model feature conjunctions up to the order of the polynomialLinear kernel allows to picking out hyper spheresRadial basis functions allow to pick out circles

7. A Hilbert space is a vector space with an inner product, which is also complete.

MaybeTrueDon't knowFalse

8. Which of the following statements is not true about kernel?

The kernel function K specify the embedding as mapping the domain set into a spaceKernel is used to describe inner products in the feature spaceThe kernel function does the mapping of the domain set into a space where the similarities are realised as outer productsThe kernel function K specify the similarity between instances

9. Which of the following statements is not true about kernel trick?

If we first map the input data into a higher-dimensional space, a linear algorithm operating in this space will behave non-linearly in the original input spaceIf the algorithm can be expressed only in terms of an inner product between two vectors, all it need is replacing this inner product with the inner product from some other suitable spaceThe mapping is always need to be computedIt provides a bridge from linearity to non-linearity to any algorithm that can expressed solely on terms of dot products between two vectors

10. Which of the following statements is not true about Kernel methods?

The data can be easily separated in the higher dimensional spaceIt can be used for pattern analysis or pattern recognitionIt maps the data into higher dimensional spaceIt only leads to finite dimensional space

11. As per the given figure Kernel trick illustrates some fundamental ideas about different ways to represent data and how machine learning algorithms see these different data representations.

TrueDon't knowMaybeFalse

12. Which of the following statements is not true about the learning algorithms?

An embedding is a way to express and utilise prior knowledge about the problem at handThe sample complexity required to learn with some kinds of kernels is independent of the margin in the feature spaceThe suitability of any hypothesis class to a given learning task depends on the nature of that taskA feature mapping can be viewed as expanding the class of linear classifiers to a richer class

13. The computational complexity challenge related to learning half-space in high dimensional feature spaces can be solved using the method of kernels.

FalseTrueMaybeDon't know

14. The Gaussian kernel is also called the RBF kernel, for Radial Basis Functions.

MaybeDon't knowTrueFalse

15. Let the domain be the real line and consider the domain points {-10, -9, -8, …, 0, 1, …, 9, 10} where the labels are +1 for all x such that |x| > 2 and 1 otherwise. The given training set is separable by a half-space.

MaybeFalseDon't knowTrue

16. The k degree polynomial kernel is defined as K(x, x^’) = (1 + ’>)^k.

FalseDon't knowTrueMaybe

17. Which of the following statements is not true about the learning algorithms?

The suitability of any hypothesis class to a given learning task depends on the nature of that taskAn embedding is a way to express and utilise prior knowledge about the problem at handThe sample complexity required to learn with some kinds of kernels is independent of the margin in the feature spaceA feature mapping can be viewed as expanding the class of linear classifiers to a richer class

18. Many learning algorithms for half-spaces can be carried out just on the basis of the values of the kernel function over pairs of domain points.

MaybeDon't knowTrueFalse

19. Which of the following statements is not true about kernel properties?

Kernel functions must be continuousKernel functions most preferably should have a positive (semi-) definite Gram matrixKernels which are said to satisfy the Mercer’s theorem are negative semi-definiteKernel functions must be symmetric

20. Polynomial-based classifiers yield much richer hypothesis classes than half-spaces.

MaybeFalseDon't knowTrue