1 Hit in 1.9 sec

PCAAE: Principal Component Analysis Autoencoder for organising the latent space of generative networks [article]

Chi-Hieu Pham and Saïd Ladjal and Alasdair Newson
2020 arXiv   pre-print
Drawing inspiration from principal component analysis and autoencoder, we propose the Principal Component Analysis Autoencoder (PCAAE).  ...  The resulting autoencoder produces a latent space which separates the intrinsic attributes of the data into different components of the latent space, in a completely unsupervised manner.  ...  In this work, we propose a network which we refer to as the "Principal Component Analysis Autoencoder" (PCAAE).  ... 
arXiv:2006.07827v1 fatcat:qxw7tmm45redvmnv2xoxdfpp3m