Dynamic, Non-Local Role Bindings and Inferencing in a Localist Network for Natural Language Understanding
Trent E. Lange, Michael G. Dyer · Neural Information Processing Systems · 1988
This paper introduces a means to handle the critical problem of nonlocal role-bindings in localist spreading-activation networks. Every conceptual node in the network broadcasts a stable, uniquely-identifying activation pattern, called its signature. A dynamic role-binding is created when a role's binding node has an activation that matches the bound concept's signature. Most importantly, signatures are propagated across long paths of nodes to handle the non-local role-bindings necessary for inferencing. Our localist network model, ROBIN (ROle Binding and Inferencing Network), uses signature activations to robustly represent schemata role-bindings and thus perform the inferencing, plan/goal analysis, schema instantiation, word-sense disambiguation, and dynamic re-interpretation portions of the natural language understanding process.