Skip to content
GitLab
  • Menu
Projects Groups Snippets
  • /
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
  • Sign in
  • Data Driven Engineering Data Driven Engineering
  • Project information
    • Project information
    • Activity
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
    • Locked Files
  • Deployments
    • Deployments
    • Releases
  • Packages & Registries
    • Packages & Registries
    • Package Registry
    • Infrastructure Registry
  • Wiki
    • Wiki
  • Snippets
    • Snippets
  • Activity
  • Graph
  • Commits
Collapse sidebar
  • cihan.ates
  • Data Driven EngineeringData Driven Engineering
  • Wiki
  • Dde 1
  • Dimensionality reduction

Dimensionality reduction · Changes

Page history
Update Dimensionality reduction authored Nov 30, 2021 by cihan.ates's avatar cihan.ates
Hide whitespace changes
Inline Side-by-side
DDE-1/Dimensionality-reduction.md
View page @ 43c328a4
......@@ -234,7 +234,7 @@ Here, U and V are [unitary matrix]( https://en.wikipedia.org/wiki/Unitary_matrix
<img src="uploads/67baf97267820063eff52b30fb4d94ca/ica_2.png" width="600">
</div>
We need to think about the last rotation: U. We need to rotate it back via $`U^*`$. How can we do that? We know that it is organized according to the variances in a hierarchical way. So I can find out on X in which direction the largest variances, i.e. the new coordinates. From the coordinates, we can find how much it was rotated ($`θ`$). The next question is; can I estimate how it may be stretched in the second step ($`Σ^{-1}`$). How was $`Σ^{-1}`$ operates in the first place? It was done according to singular values, variances. Since we can calculate the variances over the data, we can stretch it back via variances (i.e. moment). In the third step, we should again rotate back. Herein, we use another moment, kurtosis. Since we do not know how much to rotate, we will rotate in “a way that minimizes the kurtosis”. In short, we need to find the $`V`$ minimizing the kurtosis.
We need to think about the last rotation: U. We need to rotate it back via $`U^*`$. How can we do that? We know that it is organized according to the variances in a hierarchical way. So I can find out on X in which direction the largest variances, i.e. the new coordinates. From the coordinates, we can find how much it was rotated ($`θ`$). The next question is; can I estimate how it may be stretched in the second step ($`Σ`$). How was $`Σ`$ operates in the first place? It was done according to singular values, variances. Since we can calculate the variances over the data, we can stretch it back ($`Σ{-1}`$) via these variances (i.e. moment). In the third step, we should rotate back ($`V`$) . Herein, we use another moment, kurtosis. Since we do not know how much to rotate, we approximate. We will rotate in “a way that minimizes the kurtosis”. In short, we need to find the $`V`$ minimizing the kurtosis.
<div align="center">
<img src="uploads/1b71bb1a15e8c1a2e1849e8ea14a9b55/ica_3.png" width="600">
......
Clone repository
Home
  • Recommended resources
  • Toolbox
  • Miscellaneous topics
DDE 1: ML for Dynamical systems
  • Complex Systems
  • Ode to Learning
  • Regression
  • Classification
  • Clustering
  • Dimensionality Reduction
  • Outlier Detection
DDE 2: Advanced topics
  • Evolutionary learning

Imprint