File:NTK evolution.gif
Size of this preview: 800 × 533 pixels. Other resolutions: 320 × 213 pixels | 640 × 427 pixels.
Original file (900 × 600 pixels, file size: 3.96 MB, MIME type: image/gif, looped, 116 frames, 4.6 s)
 | This is a file from the Wikimedia Commons. Information from its description page there is shown below. Commons is a freely licensed media file repository. You can help. |
Summary
| Description |
English: At initialization, an ensemble of wide neural networks is a zero-mean Gaussian process; during training (gradient descent on mean-square error), the ensemble evolves according to the neural tangent kernel. The converged ensemble is a Gaussian process whose mean is the ridgeless kernel regression estimator and whose variance vanishes on the training points. |
| Date | |
| Source | Own work |
| Author | Dkarkada |
Licensing
I, the copyright holder of this work, hereby publish it under the following license:
This file is licensed under the Creative Commons Attribution-Share Alike 4.0 International license.
- You are free:
- to share – to copy, distribute and transmit the work
- to remix – to adapt the work
- Under the following conditions:
- attribution – You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- share alike – If you remix, transform, or build upon the material, you must distribute your contributions under the same or compatible license as the original.
File history
Click on a date/time to view the file as it appeared at that time.
| Date/Time | Thumbnail | Dimensions | User | Comment | |
|---|---|---|---|---|---|
| current | 17:15, 8 May 2023 | | 900 × 600 (3.96 MB) | Dkarkada | change xaxis label |
| 05:23, 3 May 2023 |  | 900 × 600 (3.94 MB) | Dkarkada | clarified NTK curve in the legend | |
| 05:14, 3 May 2023 |  | 900 × 600 (3.74 MB) | Dkarkada | Uploaded while editing "Neural tangent kernel" on wiki.riteme.site |
File usage
The following page uses this file:
