3 ms·
Train two AIs and have them check each others work. This must have been done already? No?
by partyboat1586 6y ago
Train two AIs and have them check each others work. This must have been done already? No?
- csours 6y agohttps://en.wikipedia.org/wiki/Generative_adversarial_network https://en.wikipedia.org/wiki/Generative_adversarial_network
- mcherm 6y agoAre you training the two AIs on the same data set? If so, won't they be likely to make the same sorts of errors rather than making different errors and thus providing an effective check on each other?
- partyboat1586 6y agoGood point, you would have to train them on separate data sets. I wonder how the error rate would differ from training one model on all the data.