Build Ultimate Machine Intelligence Based on the Principle of Human Brain Intelligence
Lei Tang, Shi Hui Tang · 2025
The time-space of rapid fluctuation in quantum world enables a huge amount of larger scale matter forms to have their own intrinsic attributes that also enable our brain to have the abilities of representation, memory and behavior. Originating from the subjective initiative of human, the use of tools catalyzes human natural language. The neurons easily interconnecting with each other in 3D space form the matter base of consciousness, which also has verified the fundamental viewpoint of dialectical materialism. The perception-array representation mechanism based on the radial basis function is widely used to represent color, sound, smell and gustation. This paper uses the radical basis function representation principle to derive out the approximate number of single wavelength color attributes we can perceive. It is known that the output of simple neuron in the primary cortex is sensitive to the edge orientation, this paper furtherly analyzes the attribute representation mechanism in a shape topology, then combining more shape topologies would generate the new semantics. This paper introduces that any semantics can be composed of geometric shapes and exists in a unique neuron fiber plexus node existing in an approximate zero-dimensional space. Considering the complex interconnections between semantics nodes, we introduce the semantics vector bundle structural form, in which any semantics has been defined according to the core concept of category theory. This paper also explains why our brain consumes much less power. Appendix-A introduces the core framework of building ultimate machine intelligence based on brain intelligence principle, where Relation Database with vector function plays a critical role.