Modeling human-level intelligence by integrated cognition in a hybrid architecture

Abstract

Various forms of reasoning, the profusion of knowledge, the gap between neuro-inspired approaches and conceptual representations, the problem of inconsistent data input, and the manifold of computational paradigms for solutions of these problems challenge AI models for higher cognitive abilities. We propose the I-Cog architecture as a step towards a solution for these problems. I-Cog is a modular system that is composed of a reasoning device based on analogical reasoning, a rewriting mechanism for the ontological knowledge base, and a neuro-symbolic interface for robust learning from noisy and inconsistent data.