Classification of four-qubit pure codes and five-qubit absolutely maximally entangled states
Ian Tan
TL;DR
The work addresses the classification of highly entangled quantum codes by marrying quantum information with Vinberg invariant theory. It shows that every 5-qubit AME state is LU-equivalent to a point in the unique ((5,2,3)) code C1, with equivalence governed by a finite group of order 24, and that every 4-qubit pure code is LU-equivalent to a subspace of the unique ((4,4,2)) code C2. The authors develop a Vinberg-theoretic framework to connect 4-qubit state spaces to graded Lie algebras, enabling a precise embedding and the construction of a 3-uniform infinite family of even-n states. They further provide a separating invariant set of polynomials (degrees 6, 8, 12) that fully distinguishes SU(2)⊗5 orbits of 5-qubit AME states, offering a practical criterion for equivalence. Overall, the paper advances the classification of maximally entangled states and their associated quantum codes through a synthesis of invariant theory, Lie theory, and quantum information.
Abstract
We prove that every 5-qubit absolutely maximally entangled (AME) state is equivalent by a local unitary transformation to a point in the unique ((5,2,3)) quantum error correcting code C. Furthermore, two points in C are equivalent if and only if they are related by a group of order 24 acting on C. There exists a set of 3 invariant polynomials that separates equivalence classes of 5-qubit AME states. We also show that every 4-qubit pure code is equivalent to a subspace of the unique ((4,4,2)) and construct an infinite family of 3-uniform n-qubit states for even $n\geq 6$. The proofs rely heavily on results from Vinberg and classical invariant theory.
