Regularized linear autoencoders recover the principal components, eventually

NIPS 2020, 2020.

Cited by: 4|Views34
EI
Weibo:
We investigated several algorithms that learn the optimal representation in linear autoencoders, and analyze their strength of symmetry breaking

Abstract:

Our understanding of learning input-output relationships with neural nets has improved rapidly in recent years, but little is known about the convergence of the underlying representations, even in the simple case of linear autoencoders (LAEs). We show that when trained with proper regularization, LAEs can directly learn the optimal repr...More
0
Your rating :
0

 

Tags
Comments