3 ms·
Copyright holders aren't complaining about the training per se, they are complaining about the distribution and outputs of the models which in many cases direct
by snovv_crash 3y ago
Copyright holders aren't complaining about the training per se, they are complaining about the distribution and outputs of the models which in many cases directly regurgitate the training data.
- corethree 3y agoIn most cases it won't regurgitate the same training data. What happens is the model essentially models a full on continuous curve that best fits in-between the training data. The amount of points on that curve is 9999999x more then the training data and that 999999 is not an exaggeration. It's likely too small of a number.
- snovv_crash 3y agoI disagree, the size of the models are a lot smaller than the training data. Just because I make an algorithm that linearly interpolates between two (copyrighted) values doesn't mean that it is creative or holds the wisdom between them.