Abstract of Thomas Huckle

Approximations of high-dimensional binary tensors
In quantum simulation the lowest ground state of a quantum system is related to the corresponding smallest eigenvalue of the Hamiltonian. These matrices are too large to represent even a vector of this size. Therefore, the eigenvector is represented by a matrix product tensor ansatz, which allows to solve the eigenvalue problem approximately. We describe this MPS method. Furthermore, we introduce a new approximation by representing the eigenvector as a linear combination of tensors of different block structure.


Lars Grasedyck (MPI Leipzig, Germany)
Wolfgang Hackbusch (MPI Leipzig, Germany)
Boris Khoromskij (MPI Leipzig, Germany)