2
$\begingroup$

Consider a general nonconvex QP $x^\top Qx$. This can be linearized in an extended space by using the variable $Y=xx^\top$.

Now a valid inequality $a^\top x \le b$ can be strengthend by the RLT procedure by multiplying with some $x_j$ to obtain $a^\top y \le bx_j$. See for example here.

Are there other approaches than RLT in the context of nonconvex qps to strengthen a given inequality?

$\endgroup$
2
$\begingroup$

One such paradigm is intersection cuts. It's a widely used framework in mixed-integer programming, and it naturally extends to non-convex nonlinear settings as well.

A (very) recent work on this topic: On the implementation and strengthening of intersection cuts for QCQPs.

$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.