This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
assignments:assignment1 [2016/08/24 16:11] asa [Your Report] |
assignments:assignment1 [2016/08/27 17:19] asa |
||
---|---|---|---|
Line 27: | Line 27: | ||
* $c_r = c(+1, -1)$, which is the reject cost (the cost of a false negative) | * $c_r = c(+1, -1)$, which is the reject cost (the cost of a false negative) | ||
* $c_a = c(-1, +1)$, which is the accept cost (the cost of a false positive). | * $c_a = c(-1, +1)$, which is the accept cost (the cost of a false positive). | ||
+ | |||
+ | **Incorporating the cost matrix into computing the error:** (clarification 8/27 at 5:19 pm) | ||
+ | |||
+ | The regular error | ||
+ | $$E(h) = \frac{1}{N}\sum_{i=1}^N I(h(\mathbf{x}_i) \neq y_i)$$ | ||
+ | is now replaced with: | ||
+ | $$E_{cost}(h) = \frac{1}{N}\sum_{i=1}^N c(y_i, h(\mathbf{x}_i)) \cdot I(h(\mathbf{x}_i) \neq y_i)$$ | ||
With these definitions, answer the following: | With these definitions, answer the following: |