An IQP Implementation

The IQP Algorithm

The IQP Algorithm minimizes the function $f(x)=(1/2)xTQx+cTx$ subject to the general linear equality constraints $A$₁x = b₁ and the general linear inquality constraints $A$₂x >= b₂ and the simple lower and upper bound constraints $l$_i <= x_i <= u_i for i = 1,2,...,n

The matix: is an mxn matrix and the vector: is an m vector. The symmetric matrix Q need not be positive definite. If Q is not positive definite, only a local minimum may be found.

Available Reference Information

• The IQP -- quadratic programming reference page

Available Software

• double precision implementation -- dall.f
• single precision implementation -- dall.f