Request Confirmation Networks for Neuro-Symbolic Script Execution

We propose Request Confirmation Networks (ReCoNs) to combine neural learning with the sequential, hierarchical detection of sensory features, and to facilitate planning and script execution. ReCoNs are spreading activation networks with units that contain an activation and a state, and are connected with typed directed links that indicate partonomic relations and spatial or temporal succession. By passing activation along the links, ReCoNs can perform both neural computations and controlled script execution. We demonstrate an application of ReCoNs in the cognitive architecture MicroPsi2 for an active perception task in a virtual environment.