Tikhonov regularization (under construction)

Summary

-No Summary-

Notes

This concept is still under construction.

Context

This concept has the prerequisites:

learning invariances in neural nets (Tikhonov regularization is a way of learning invariances in neural nets.)

See also

Tikhonov regularization is one strategy for learning invariances in neural nets. Others include:
- building it explicitly into the architecture, as in convolutional nets
- augmenting the training set with warped examples
- tangent propagation , which penalizes instability with respect to transformations
Denoising autoencoders are a neural net architecture based on the idea of invariance to noise in the inputs.