Cognitive AI focuses on functionally modelling human memory, reasoning and learning inspired by the evolution of neural systems over 500 million years, and decades of work in the cognitive sciences. This talk will introduce work on cognitive architecture, the chunks graph data and rules language, and its application in a series of web-based demos, including smart homes, industrial robots, and ongoing work on natural language. Semantics are expressed in terms of cultural norms, following Philip Johnson-Laird who noted that humans don’t rely on the laws of logic and probability, but rather by thinking about what is possible. Future work will seek to integrate emotions and cognition. This is very much in contrast to the Aristotelian tradition of logic, formal semantics and model theory.

Slides fro the seminar: https://www.w3.org/2020/CogAI-2020-11-06.pdf