next up previous
Next: Up: Previous:

Overfitting


Consider error of hypothesis h over

Hypothesis $h \in H$ overfits training data if there is an alternative hypothesis $h' \in H$ such that

errortrain(h) < errortrain(h')

and

\begin{displaymath}error_{\cal{D}}(h) > error_{\cal{D}}(h') \end{displaymath}