Calculating method based on brain-like polysensory attention switching

Abstract

The invention belongs to the artificial intelligence and cognitive neuroscience integration field and especially relates to a calculating method based on brain-like polysensory attention switching. The method is used to solve the problem of reliable information selection in a multi-sensory information input environment. The method comprises the following steps of S1, based on a cerebral visual cortex model, carrying out the content identification of a digital image and acquiring a visual digital sequence; S2, based on a cerebral auditory cortex model, carrying out the content identification ofa digital audio and acquiring an auditory digital sequence; S3, based on the digital sequences, using a digital inference model to carry out digital inductive inference, and calculating and storing arule between the digital sequences; and S4, based on a visual-auditory attention switching model, selecting information with a high weight as current reliable modal information and carrying out inference calculation, and acquiring an identification result. In the invention, a series of humanoid behaviors such as vision, an auditory sense, inference, attention switching and the like can be completed simultaneously, and under the different environments, the reliable information is accurately selected for further processing.

Description

technical field [0001] The invention belongs to the comprehensive field of artificial intelligence and cognitive neuroscience, and specifically relates to a calculation method based on brain-like multisensory attention switching. Background technique [0002] With the development of the field of artificial intelligence, academia and industry have higher and higher requirements for the intelligence of machine systems, and hope that they can reach the level of brain information processing to a certain extent. So far, how to make computers have a higher level of intelligence and stronger environmental adaptability is still a very challenging problem. [0003] In 2013, the proposal of the EU Brain Project opened up new ideas for the field of artificial intelligence. The plan studies the information processing mechanism of the brain through different scales of micro, meso and macro, and applies it to the field of intelligence science. Ultimately, it hopes to build Create a gener...

AI Technical Summary

Problems solved by technology

Among them, multimodal information suppression and switching has a certain research basis in the field of neuroscience, but less research in the field of artificial intelligence

In fact, with the development of artificial intelligence systems, the input of multi-sensory information has become inevitable, but how to select reliable information for further processing in different environments is a difficult point

Images