Skip to content

Rohit01-zoey/Perturbation-techniques-in-CNNs

Repository files navigation

Perturbation-techniques-in-CNNs

Backpropagation is a technique used in machine learning models which is fast and powerful. However it usually requires very precise calculation power which is not always possible.

To solve this constraint, we implement perturbation techniques in our CNN model. This includes weight perturbation and node perturbation. These techniques have been found out to be slower however use up very less computing power. As a result, these can be implemented on-chip.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published