5,337
edits
Line 46: | Line 46: | ||
===Bias-Variance Tradeoff=== | ===Bias-Variance Tradeoff=== | ||
* Let <math>L_D(h)</math> be the true loss of hypothesis h | |||
: and <math>L_S(h)</math> be the true loss of hypothesis h | |||
* <math>L_D(h_s^*) = L_D(h_D^*) + [L_D(h_s^*) - L_D(h_D^*)]</math> | |||
* The term <math>L_D(h_D^*)</math> | |||
* The term <math>[L_D(h_s^*) - L_D(h_D^*)]</math> is called variance. | |||
* Larger hypothesis class will get smaller bias but larger variance. | |||
* Overfitting vs. underfitting |