Toward a theory-based natural language capability in robots and other embodied agents: Evaluating Hausser's SLIM theory and database semantics

Computational natural language understanding and generation have been a goal of artificial intelligence since McCarthy, Minsky, Rochester and Shannon first proposed to spend the summer of 1956 studying this and related problems. Although statistical approaches dominate current natural language applications, two current research trends bring renewed focus on this goal. The nascent field of artificial general intelligence (AGI) seeks to evolve intelligent agents whose multi-subagent architectures are motivated by neuroscience insights into the modular functional structure of the brain and by cognitive science insights into human learning processes. Rapid advances in cognitive robotics also entail multi-agent software architectures that attempt to parallel in many ways the sensory and cognitive processes of humans. Natural language capability is a key objective for both types of software, whether embodied in a physical robot or in a virtual world that emulates features of the physical environment.Hausser's SLIM theory of natural language communication and associated Database Semantics computational instantiation are an ambitious attempt to bridge the gap between formal theory approaches to computational natural language capability and an embodied approach to language and meaning which requires integration of language with sensory perception, planning and social interaction. This dissertation evaluates Hausser's approach to the development of human-level computational natural language capability in embodied and socially situated agents and argues that a theoretical basis for such capability is emerging as a result of recent evidence from linguistics, cognitive science and neuroscience.