Hybrid techniques to break down large graphs into smaller graphs
With the current D-Wave hardware, a K65 graph is the largest graph that has a minor embedding that can be mapped onto the 2000Q chip. To solve large problems with more than 65 nodes, hybrid classical-quantum techniques have to be used.
Are there hybrid techniques that systematically allow to obtain approximate solutions to large graph problems, by either splitting the original graph into smaller subgraphs and then by combining the results of the anneals on the small subgraphs, or by reducing the size of the original graph down to to a K65? If so, is it possible to use the Ocean SDK to implement such hybrid solutions?