A (primal) hard-margin SVM is an optimization problem of the following form: $\min\limits_{\alpha_1,\ldots,\alpha_n\geq 0} \frac{1}{2} \sum \limits_{i,j = 1}^n \alpha_i \alpha_j y_i y_j k(x_i, x_j)$ subject to $y_i f(x_i) \geq 1, i = 1, \ldots, n$ where $f(x) := \sum_{i=1}^n \alpha_i y_i k(x_i, x)$