I am looking at Autoencoders in keras. They say,
"Autoencoding" is a data compression algorithm where the compression and decompression functions are 1) data-specific, 2) lossy, and 3) learned automatically from examples rather than engineered by a human
My assumption is, because it is lossy, using Autoencoder does not increase the accuracy. Say, I have a seq2seq model for text summarization, if I use Autoencoder I can speed up the training process but it will not increase the accuracy of the model. Is this assumption correct?