September 2018
Intermediate to advanced
296 pages
9h 10m
English
The last step of the pipeline involves training a value network to evaluate the board state, or in other words, to determine how favorable a particular board state is for winning the game. Formally speaking, given a particular policy,
, and state,
, we would like to predict the expected reward,
. The network is trained by minimizing the mean-squared error (MSE) between the predicted value, , and the final outcome:
Where are the ...
Read now
Unlock full access