3 ms·
Everything you said doesn't dispute the above comment and agrees with its core premise: "In terms of model improvement, yes synthetic data can help. In terms o
by missosoup 7y ago
Everything you said doesn't dispute the above comment and agrees with its core premise:
"In terms of model improvement, yes synthetic data can help. In terms of the arms race? No. True examples provide knowledge that is unique. "
- EsssM7QVMehFPAs 7y agoI was rather commenting on the first part implying that training a neural network with the statistical distribution that comes out of a GAN or VAE does not add value beyond that generative model capabilities. I do not agree on that because as I explained, with domain knowledge it is very much possible to shape the data generated for augmented learning - beyond the plain statistical variations of GAN and similar, which are obviously of very limited value in training.