I have the following problem to optimize:
\begin{align} \begin{split} \hat{\mathbf{x}}, \hat{\delta} = \underset{\mathbf{x}, \,\delta}{argmin} \, \sum_{i=1}^l \, \sum_{j=1}^m \, a_{i,j}\left(x_i + \delta \right)^2 \\ s.t. \, \, \, \, 1 - x_i \leq 0 \, \, \forall \, i \\ \text{and} \,\, \delta \leq 0 \end{split} \end{align} where $a_{i,j}$ is just a variable that doesn't depend of $x_i$.
I am thinking to solve this problem by minimizing the following Lagrangian function: \begin{align} \underset{\mathbf{x}, \, \delta}{argmin} \, L\left(\mathbf{x}, \, \delta, \, \lambda_1, \, \lambda_2\right) = \sum_{j=1}^m \, \sum_{i=1}^l \, a_{i,j}\left(x_i + \delta \right)^2 + \lambda_1 \left(1 - x_i\right) +\lambda_2 \, \delta \end{align} and then find the derive =0.
I am sure I am wrong. Can someone provide me with brief mathematical details about how this problem can be solved?
Any help will be very appreciated.