This invention describes a computer method to efficiently match things that need resources (like computing jobs or teams) with available resources (like processors or shared workspaces). It starts by arranging the things needing resources on a virtual map using a physics-like simulation. This map then guides an initial setup on a second, more complex graph. Finally, the things needing resources virtually move around this second graph, expanding until they find and claim the necessary resources. The claims specifically focus on matching entities where the second entity provides physical, on-demand shared resources in a workspace.
Why it matters: The invention uses a graph-based search for resource allocation. Since 2019, the field of graph neural networks (GNNs) has advanced significantly, potentially offering more sophisticated and efficient ways to implement the 'neural graph' search and optimization described, especially for dynamic resource environments.
AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.
Reinvent this with AI