| Explanation-based Neural Networks (EBNN) are powerful inductive learning systems which can utilise background knowledge in addition to empirical data. However, EBNNs are based on simple multi-layer perceptrons (MLPs) and have limited knowledge representation capacity. Connectionist knowledge representation systems are now available and can be used to extend EBNN. We combine EBNN and connectionist knowledge representation to arrive at connectionist analytical learning. | |
Add a project to the database.
Update the entry for this project.