coolrunning / monogamy Goto Github PK

Figuring out additional jumping points of werner states coexistence.

Strictly spoken, we can have complex values in the density matrices. It needs to be evaluated, whether this is possible in conjunction with CVXPY.

There is an analytic formula for the experiment done in that notebook.

There are some simple checks whether two substates can have an extension. This could be employed before constructing the optimization problem.

The bitmaks to mark the traced out dimensions should be given alternatively as a list of indices describing the corresponding subsystems.

Currently the monogamy uses a function to translate flat indices into multi-indexes. Numpy offers this functionality by having arrays of higher dimensions. This could simplify a lot of code.

As an additional familiy of states, Bell diagonal states were suggested to explore.

Extend monogamy to work together with Qutip.

Implement a generator that produce random density matrices.

Too high precision around the jumping points of the werner states causes crashes in CVXPY which need to be investigated.

So far, constraints need to be parsed elementwise for CVXPY, which could be rather inefficient for large matrices.