Neil Dowling

---

The process tensor framework provides an operational approach for describing non-Markovian open quantum systems, allowing the encoding of arbitrary multi-time observables and providing unambiguous notions of non-markovianity. This offers advantages in areas ranging from classical simulation of open systems to device characterization and error mitigation. However, for complex dynamics, the computational requirements of simulating a process tensor scale poorly using standard matrix product operator techniques. After introducing the process tensor framework, I will describe our new ansatz of process tensors based on a tensor-tree geometry, and show that they are particularly well-suited for describing strongly correlated systems with long-range memory, such as the paradigmatic spin-boson model. Beyond this, leveraging 2D tensor network renormalization group methods, I will describe our algorithm for deriving a process tree from an underlying Hamiltonian, via the Feynmann-Vernon influence functional. Our work lays the foundation for the development of more efficient numerical techniques in the field of strongly interacting open quantum systems.

---

University of Cologne
0.03
Contact: Silvia Pappalardi