Performance comparison of Gilbert’s algorithm and machine learning in classifying Bell-diagonal two-qutrit entanglement

While classifying states as entangled or separable is one of the fundamental tasks in quantum information theory, it is also extremely challenging. This task is highly nontrivial even for relatively simple cases, such as two-qutrit Bell-diagonal states, i.e., mixtures of nine mutually orthogonal maximally entangled states. In this article we apply Gilbert’s algorithm to revise previously obtained results for this class. In particular we use “entanglement cartography” to argue that most states left in [Hiesmayr, B. C. Scientific Reports 11, 19739 (2021)] as unknown to be entangled or separable are most likely indeed separable, or very weakly entangled, beyond any practical relevance. The presented technique can find endless applications in more general cases.