Sum of squares¶
\[RSS(w) \triangleq \sum_{i}^N (y_i - \hat{y}_i)^2\]
\(\hat{y}_i\) is the predicted value of i
It can be viewed as the \(l_2\) of error:
\(RSS(w) = ||\epsilon||_2^2 = \sum_{i=1}^N \epsilon_i^2\)
\(\epsilon_i =y_i - \hat{y}_i\)
\(\hat{y}_i\) is the predicted value of i
It can be viewed as the \(l_2\) of error:
\(RSS(w) = ||\epsilon||_2^2 = \sum_{i=1}^N \epsilon_i^2\)
\(\epsilon_i =y_i - \hat{y}_i\)