Question $4[5 \%]$ Referring to the ridge regression problem $$ \min _{w \in H} \frac{1}{2 n} \sum_{i=1}^{n}\left(y_{i}-\left\langle w, \Lambda\left(x_{i} ight) ight angle ight)^{2}+\frac{\lambda}{2}\|w\|^{2}, $$ and denoting by $\mathrm{K}$ the Gram matrix of the data, indicate the solution of the dual problem a) $\bar{u}=(\mathrm{K}+\lambda n \mathrm{~d}) y$ b) $\bar{u}=\left(\mathrm{K}^{2}+\lambda n \mathrm{ld} ight)^{-1} y$ c) $\bar{u}=(\mathrm{K}+\lambda n \mid \mathrm{d})^{-1} y$.