Kernel trick refers to kernel methods in machine learning, such as kernel support vector machine (SVM) or kernel principal components analysis (PCA). It allows to generalize linear techniques to non-linear situations. DO NOT USE this tag for [kernel] which is reserved for non-parametric estimation ...

learn more… | top users | synonyms

1
vote
0answers
9 views

Combining text and non-text features

I am working on a binary classification problem using SVM. I am currently using ksvm in R (kernlab package). The input is a combination of text and scores. I would like to be able to use substring ...
10
votes
3answers
248 views

How to intuitively explain what a kernel is?

Many machine learning classifiers (e.g. support vector machines) allow one to specify a kernel. What would be an intuitive way of explaining what a kernel is? One aspect I have been thinking of is ...
0
votes
0answers
17 views

Combination of two SVM Kernels

According to the book "Support Vector Machines" from Cristianini and Shawe-Taylor, it is feasible to make kernels from kernels. My question is now more in application of this methods with tools like ...
0
votes
1answer
20 views

Prove that a kernel is conditionally positive definite

A kernel is called positive definite (p.d) if its Gram matrix is p.d., i.e. all eigenvalues of the Gram matrix are positive for all possible input vectors in the feature space. My understanding of ...
4
votes
1answer
49 views

Is Support Vector Machine sensitive to the correlation between the attributes?

I would like to train an SVM to classify cases (TRUE/FALSE) based on 20 attributes. I know that some of those attributes are highly correlated. Therefore my question is: is SVM sensitive to the ...
0
votes
0answers
13 views

Neural networks kernel for high error rate in training set.

I will be working with a huge training set (around 10^6 examples, with around 400 features). Which has labels (around 100) accurate to around 90%. It would be possible to generate a smaller subset of ...
1
vote
0answers
34 views

Generalization error of PCA and kernel PCA

I've been recently reading Shawe-Taylor et al. 2005, On the Eigenspectrum of the Gram Matrix and the Generalization Error of Kernel PCA, where the authors analyze the squared residual of kernel ...
4
votes
1answer
67 views

Using kernels with Fisher's linear discriminant analysis

I am a bit stuck implementing the Kernel Fisher Discriminant. $$ J(\mathbf{w}) = \frac{\mathbf{w}^{\text{T}}\mathbf{S}_B^{\phi}\mathbf{w}}{\mathbf{w}^{\text{T}}\mathbf{S}_W^{\phi}\mathbf{w}} $$ $$ ...
1
vote
2answers
72 views

SVM kernel parameters value [duplicate]

I need a kernel for the following situation: 100 dimensions, 10 classes For every feature(in the features order) the maximum distance between any different pair of points is ...
2
votes
1answer
31 views

Turn a distance measure into a kernel function

I have read here that an easy way to turn a distance function $d$ into a similarity function $s$ is to compute: $s = e^{-\gamma * d}$. I believe that this is also what is done with the RBF kernel. ...
1
vote
1answer
33 views

What is the nonlinear transformation assumed by the gaussian (rbf) kernel?

A common kernel choice is the gaussian kernel: $ k(x,x^{'}) = \exp \big( -\frac{1}{2\sigma^2}\| x - x^{'} \|^2 \big)$ This implies a transformation on $x$, and equally on $x^{'}$. What is it?
0
votes
1answer
12 views

Kernel SVM overfitting after training set extension

I am training Kernel SVM from sklearn package for binary classification problem. I perform a gridsearch for parameters optimization. Parameters are taken from following ranges: 'C':[1., 10., 100.], ...
0
votes
1answer
42 views

Does the kernel trick really map 2d data to 3d data?

I want to learn something about kernel trick in svm, so I'm using this code: ...
0
votes
0answers
6 views

What are “parts” in Haussler's definition of R-convolution kernels?

I have been reading about R-convolution kernels: http://citeseerx.ist.psu.edu/viewdoc/download?rep=rep1&type=pdf&doi=10.1.1.110.638. These important types of kernels are generalization of ...
3
votes
1answer
136 views

Confused about transposes in kernel notation

I am studying machine learning and I ran into a challenge that does not make sense to me. Maybe it is a crazy or a simple question. If we have kernel $ \phi_1(x)=[x, x^2]^T $ and $ \phi_2(x)=[2x, ...
1
vote
0answers
26 views

Probabilistic degree of confidence for the kernel SVM with RBF

Let $f\colon\Bbb{R}^n\to\Bbb{R}$ be the decision function of an SVM using the radial basis function (RBF), $$ k(\mathbf{x},\mathbf{x}')=\exp\Big(-\gamma\|\mathbf{x}-\mathbf{x}'\|^2\Big). $$ That is, ...
0
votes
1answer
56 views

Locally weighted regression VS kernel linear regression?

I am trying to make it clear the relationship of the listed three methods. According to my understanding kernel regression means : the weight vector W lies in the space spanned by training data. $$ ...
1
vote
1answer
58 views

Kernel methods in machine learning?

I am beginning to tackle geostatistics problems where I tried to apply kriging(gaussian processes) to interpolate demographical water drop. According to my understanding, kernel methods are something ...
3
votes
1answer
68 views

How to use the kernel trick on data that can't be visualized

I'm reviewing the kernel trick and there are a lot of toy examples of how a 2D classification which can't usually be separated by a linear SVM can be separated in 3 space. This is fine, but how is ...
1
vote
2answers
95 views

Kernel PCA and classification

I need to perform kernel PCA on the colon-­‐cancer dataset and then I need to plot number of principal components vs classification accuracy with PCA data. For the first part I am ...
0
votes
2answers
30 views

Data normalization for RBF kernel

I have a matrix of values where rows are individuals and columns are attributes. I want to extract a similarity value for every pair of individuals, and I use an rbf kernel: $$k(x_i,x_j) = ...
3
votes
1answer
71 views

How do you transform a decision boundary in the angle kernel to the original space?

Say I have training data $S_n$ and each point is of the form $x = \langle x_1 , x_2 \rangle$ in the original space (i.e. $x^{(i)} \in \mathbb{R}^2$). I was considering the following kernel: $$ ...
2
votes
0answers
40 views

How to prove that Radial Basis Function can be derived by mapping function?

How to prove the radial basis function $k(u,v) = \int_{\mathbb{R}^d} \phi_t(u)\phi_t(v)dt $ can be integrated out by mapping function? $$\phi_{t}(u) = \frac{1}{(2\pi\Sigma)^{d/2}} ...
0
votes
0answers
37 views

SVM for an unbalanced textual dataset?

I have a text classification task, currently I can classify the data with very poor precision. This are the scores: ...
1
vote
1answer
36 views

Regularizing the dual variables in SVM

Consider the the optimization program of the kernelized SVM: $$\text{maximize}_{\alpha} ~~ \alpha^T1-\alpha^TQ\alpha$$ $$\text{subject to:} \sum_{i=1}^N \alpha_iy_i=0,~0\leq \alpha_i\leq c$$ where ...
0
votes
0answers
51 views

What are the potential disadvantages of doing kernel PCA?

I was trying to learn more of the motivation around kernel PCA. Its clear to me that one might need to change the representation of the data if it lies in a non-linear space, hence, the projection ...
4
votes
1answer
43 views

Does a polynomial kernel with degree less than 1 satsify mercers condition

Consider the polynomial kernel: $$K(\boldsymbol{x}, \boldsymbol{x}') = (\boldsymbol{x}^{T} \boldsymbol{x}'+c)^{d}$$ This kernel satisfies the mercers theorem/condition. Since I never saw any ...
0
votes
1answer
133 views

Which hyperplane separates these two classes?

I have a dataset of 3 dimensional points in two classes, I want to separate between the two. As the plot suggests, these two are completely separable but I don't know the formula to form the ...
4
votes
1answer
185 views

How to choose a kernel for kernel PCA?

What are the ways to choose what kernel would result in good data separation in the final data output by kernel PCA (principal component analysis), and what are the ways to optimize parameters of the ...
3
votes
1answer
84 views

What exactly is the procedure to compute principal components in kernel PCA?

In kernel PCA (principal component analysis) you first choose a desired kernel, use it to find your $K$ matrix, center the feature space via the $K$ matrix, find its eigenvalues and eigenvectors, then ...
1
vote
1answer
49 views

Integrating length for input-space feature PC projections in kernel PCA

I read a paper detailing the algebraic process of kernel PCA. I have question though: the paper details the projection of new points onto the new eigenvectors in the feature space, but what if I want ...
18
votes
4answers
547 views

What makes the Gaussian kernel so magical for PCA, and also in general?

I was reading about kernel PCA (1, 2, 3) with Gaussian and polynomial kernels. How does the Gaussian kernel separate seemingly any sort of nonlinear data exceptionally well? Please give an intuitive ...
1
vote
1answer
53 views

What are the various basic kernels available?

I am currently following the book Gaussian Processes for Machine Learning by C.E. Rasmussen and C.K.I. Williams and I have come across various kernels in their Chapter 4 I have also gone through the ...
1
vote
0answers
39 views

How to fit a single quadratic term to a regression

I have a high dimensional multivariate model and am fitting linear weights to each of the $N$ free variables using a classic stable SVD matrix solver. This works. I want to improve the fit by using a ...
1
vote
0answers
13 views

U statistics and RKHS

Is there a relation between the kernel function in U-statistics and RKHS theory? Namely, can the kernel trick be seen as an order-2 U-statistic?
1
vote
1answer
80 views

Is it possible to project a new vector onto the PC space using kernel PCA?

Let $X_{N \times d}$ be the data matrix, where $N$ is the number of samples and $d$ the size of the features space. Using kernel PCA (kPCA), one first computes a kernel matrix $K_{N \times N}$, and ...
1
vote
1answer
50 views

Intuition behind RKHS

Why has RKHS become such an important concept in machine learning in recent times. Is it because it allows us to represent a function of combination of linear functions? What areas of mathematic does ...
1
vote
0answers
48 views

Non-decaying eigenvalues in Kernel PCA with small kernel width

I noticed that when I use a small width kernel (RBF) with PCA, I get my desired result (clustering in this case), but I do not get a decay in the eigenvalues (they stay about the same value). Is that ...
0
votes
0answers
17 views

Kernel Methods for Binary Vectors

I am currently involved in a project which requires a minor point in choosing a proper similarity metric for a set of binary vectors, i.e. all components are either 1 or 0 . Currently, the go-to ...
2
votes
2answers
59 views

Kernel methods on Categorical Data

I have a basic understanding of kernel methods and the kernel-trick and the advantages of it, why it is preferred over conventional machine learning algorithms etc. However, I have some trouble using ...
0
votes
0answers
23 views

Pyramid Match Kernel: How to fit histogram grid around data points?

This question is directly related with Kernel methods and SVM, so I think this is a good place to ask it. I am planning to use Pyramid Match Kernel method for object recognition from depth images: I ...
2
votes
1answer
89 views

How does the $\phi(x_i)$ function look for Gaussian RBF kernel?

I am trying to write programs for simple SVM cases. And what I am stuck at is that I am unable to find $\phi(x_i)$ functions for given kernels. For example there is Gaussian Radial Basis Function ...
0
votes
0answers
41 views

Support Vector Machine with zero bias term

I'm looking for an algorithm to solve SVM with zero bias term. So dual form of such SVM is $max_\alpha \sum_i^n \alpha_i -1/2\sum_i^n \sum_j^ny_iy_jK(x_ix_j)\alpha_i\alpha_j$ subject to: $0 \leq ...
0
votes
0answers
70 views

The Shogun Machine Learning toolbox for SVM with precomputed kernel and zero bias

Can I use the Shogun Machine Learning toolbox for SVM with precomputed kernel and zero bias. I should be able to input pre-computed kernel and I also should be able to set bias zero.
0
votes
0answers
71 views

SVM with pre-computed kernel and zero bias

I have an optimization function, where I need to give my own kernel matrix and bias value is zero. The kernel matrix is calculated using the data but there is no specific formula for it. If I have a ...
0
votes
0answers
69 views

LIBSVM for pre-computed kernel and zero bias (b values is zero)

I want to do binary classification and I'm using LIBSVM library for that. I have a precomputed Kernel and my bias value (b) is zero. Can I do this in LIBSVM or do I have to use some other library? ...
0
votes
0answers
34 views

Usage of libsvm with RBF kernel and no Offset

I'm using libsvm for the binary classification and using a precomputed Kernel. In my particular problem there is no bias term (it's zero). Is there anyway to adjust the bias term in libsvm (and not ...
0
votes
1answer
79 views

What is the toolkit that implements Cost sensitive Support Vector Machine?

I need implementation of cost sensitive support vector machine. The cost is different for each training example (unlike each class). So problem is to solve $max_\alpha$ $-1/2 \sum_{i,j} ...
1
vote
1answer
131 views

Which PCA (or kernel PCA) basis better describes a single test sample?

I have two PCA bases obtained by decomposition of two groups of training data. I also have some samples of test data. How can I decide which PCA basis fits better each test sample? I tried to ...
0
votes
0answers
16 views

definig distances using radial basis functions

in svm the kernels are supposed to measure the distance between two vectors in the feature space. however, rbf is largest at 0 meaning that in that new space the distance between a feature and itself ...