NQCP Life in Quantum - Lead for Quantum Technology Development Kristin Björg Arnardóttir , MQS

More chemistry for less compute: Tensor networks and quantum algorithms in industry

At this talk, I will start by giving a general overview of what we do at Molecular Quantum Solutions (MQS), and will then go on to present two of our methodological advances that enhance practical quantum algorithms for molecular electronic structure calculations. First, I will introduce a hybrid tensor network-variational quantum eigensolver (TN-VQE) approach that addresses VQE convergence challenges. By encapsulating the Hamiltonian into a Matrix Product Operator within a parameterized unitary tensor network, we partition optimization between classical tensor network methods and quantum circuits, demonstrating improved convergence while maintaining shallow circuit depths for NISQ devices. Second, I will describe optimizations to quantum phase estimation (QPE) that dramatically reduce qubit requirements. I'll shown that using a control state of an MPS representation of The Discrete Prolate Spheroidal Sequence (DPSS) state greatly increases the accuracy, even with low bond dimensions, and how the MPS nature of the state allows us to recycle the necessary qubits.
At MQS we develop a software platform which integrates quantum chemistry, machine learning, and quantum computing methods for molecular simulation and property prediction. We work on problems spanning pharmaceutical formulation, process optimization, and materials design; and we recently moved offices to Quantum Denmark!