When dealing with Linear regressin with dependent variables, one can consider the optimization problem: $$\arg \min_w 0.5\lVert{w}\rVert^2 \\ s.t. Xw=y$$

Where $X\in\mathbb{R}^{n,d}$ is the data matrix and $y\in\mathbb{R}^{n}$ is the labels vector.
Using Lagrangian multipliers I was able to show that $w^*$ can be written as $w^*=X^T\alpha$ for $\alpha \in \mathbb{R}^n$.

Now given a new $x\in\mathbb{R}^d$, consider $x^Tw^*$. How can this be expressed using dot products between $x\in\mathbb{R}^d$? The goal is to show that the kernel trick is working for that setup of linear regression.



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.