David Shin
David Shin
Everywhere else, as far as I can tell, the "E" in the acronym NNUE stands for the word "efficiently" ([example](https://en.wikipedia.org/wiki/Efficiently_updatable_neural_network)). But in this book, the word "effectively" is used.
How difficult would it be to create the empty board correlation map directly from self play ownership data, without the network?
I know that some of the self-play games are handicap games. Are these games included for correlation training? If so, might they impact empty board maps?