Speaker
Description
While deep learning based systems have recently produced significant advances in their capability of authoring sensible text, some researchers argue that a different approach is needed to achieve true language understanding. I will present the Concept Network Model of human thinking and comprehension, which is a hybrid symbolist-connectionist approach. This model draws its motivation not from the biology of the brain (like neural networks do) but from cognitive psychology, which is the proper level to identify the foundations of thinking. The model is implemented in the Cogito system, which is capable of the basic processing steps of human comprehension (like spreading activation, priming, disambiguation, coreference resolution, mapping the current thought into earlier knowledge, context dependent interpretation, and reasoning). I will introduce the Human Mind Project, a brand new initiative that aims to achieve human level language understanding via a global cooperation of researchers on the www.humanmind.uk portal.