The intended quantity is the <Leave-one-out cross-validation> error
$$
\operatorname{LOOCV}
=\frac1n\sum_{i=1}^n
\ell\!\left(\widehat h^{(-i)}(X_i),Y_i\right),
$$
where $\widehat h^{(-i)}$ is trained without observation $i$ and the loss is the zero-one misclassification loss. Computing it literally requires fitting $10000$ neural networks, which is prohibitively expensive.

Solution

Past exam of the mathematics course of the University of Cambridge / 2025 / iii / Paper 218 / 4 / e / Solution

 Solution

ID: past-exam-of-the-mathematics-course-of-the-university-of-cambridge/2025/iii/paper-218/4/e/solution