A scalable 2-local architecture for quantum annealing of all-to-all Ising models

Achieving densely connected hardware graphs is a challenge for most quantum computing platforms today. This problem is particularly crucial for quantum annealing (QA), a lesser known flavour of quantum computation with respect to the digital model. In this talk, we’ll first motivate the need for dense connectivities in the context of QA by introducing this computational framework and its standard use cases. We will then present our proposed architecture, a resource-efficient configuration that effectively realises an Ising Hamiltonian with all-to-all connectivity within a graph of degree d=3 containing exclusively 2-local, Ising interactions. By describing the original problem graph in terms of triangles, we are able to derive this maximally sparse configuration based on chains of qubits encoding logical variables. We thus show an alternative route to scale up devices dedicated to classical optimization tasks within the quantum annealing paradigm.

Presential in the seminar room. Zoom link:

https://zoom.us/j/98286706234?pwd=bm1JUFVYcTJkaVl1VU55L0FiWDRIUT09