Questions tagged [kernel]

A kernel is a function that acts as a parameter that is passed to some algorithm. For example, the two dimensional Gaussian kernel is often used as a parameter for low pass filtering in image processing.

5 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
1
vote
0answers
64 views

Difference in the meaning of kernel, filter, and dictionary atoms

While reading literature on wavelets, I have encountered keywords filters and dictionary atoms and their exact meaning is confusing for me. Basically, one can use them interchangeably and I won't see ...
0
votes
0answers
11 views

KCS - 3D profile representation

I was reading Character Recognition Systems book and in section 2-3-2-3 a new kind of filter is introduced (instead of normal Gaussian filter) to reduce mask size and to prevent data loss by the name ...
0
votes
0answers
12 views

Compute centered kernel value after zero-centering a kernel matrix

I am stuck trying to implement denoising with Kernel PCA. Most kernel is not supposed to be zero-mean. Say we have training inputs $[{\bf x}_1,\cdots, {\bf x}_n]$. The kernel (Gram) matrix is then $$ ...
0
votes
0answers
20 views

How would I modify a squared-exponential covariance kernel to be periodic?

I want to create a Gaussian process that resembles one generated by an exponential kernel (smooth and with a lot of variance) with a single caveat: I need the final value to be the same as the initial ...
0
votes
0answers
50 views

Problem implementing KL expansion with square-exponential kernel: output just looks like Gaussian

I am trying to implement the Karhunen Loeve expansion for a 1-D Gaussian random field with a square-exponential kernel. Specifically, I know that a Gaussian process has a KL expansion $\hat{U}=\sum_{...