Obtaining Training Examples

• V(b): the true target function
• $\hat{V}(b)$ : the learned function
• V_train(b): the training value

One rule for estimating training values:

• $V_{train}(b) \leftarrow \hat{V}(Successor(b))$