2 ms·
This is fundamentally why we have deep neural networks. Fitting a simplified curve in a huge space. The combination you named may never have been observed befo
by tomrod 8y ago
This is fundamentally why we have deep neural networks. Fitting a simplified curve in a huge space.
The combination you named may never have been observed before. The curse of dimensionality strikes.
- graycat 8y agoYup. Did I mention that with cross tabulation we'd need a lot of data!!!! :-)!!!
- tomrod 8y agoYou did!
- AnthonyMouse 8y agoBut isn't that the whole problem? You quickly run into the case where the required sample size is larger than the entire population size.
- graycat 8y agoSo, you expect us to boil it all down to just F = ma, and f'get about the apple, the apple tree, the time of day, the phase of the moon, the temperature of the day, what Newton was wearing????? That was a really good shot at causality. Let me just say, quite generally, causality is super tough to find.