In the process of preparing pull request on MIQCQP functionality, I noticed that the factor 1/2 derived from evaluating the quadratic term $x^{\top}Hx$ at the given value of selected variables could have been omitted in miqps_master() when computing $c$ and the constant value of the objective. In particular, I think the following changes are required:
- Line 261:
cc = cc + 1/2 * (H(~j,j)+H(j,~j)') * x(j);
- Line 268:
f = f + 1/2 * x(j)' * H(j,j) * x(j);
Am I right?
In the process of preparing pull request on MIQCQP functionality, I noticed that the factor 1/2 derived from evaluating the quadratic term$x^{\top}Hx$ at the given value of selected variables could have been omitted in $c$ and the constant value of the objective. In particular, I think the following changes are required:
miqps_master()when computingcc = cc + 1/2 * (H(~j,j)+H(j,~j)') * x(j);f = f + 1/2 * x(j)' * H(j,j) * x(j);Am I right?