Unlock instant, AI-driven research and patent intelligence for your innovation.

A measure of intelligence for non-human agents

A measurement method and technology of agents, applied in the fields of instrumentation, error detection/correction, calculation, etc., can solve the problem that EM evaluation method is not suitable for the field of image recognition, and evaluate the intelligence of agents.

Active Publication Date: 2020-10-20
INST OF COMPUTING TECH CHINESE ACAD OF SCI
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, there is currently no general formula to measure the intelligence of an agent in various aspects. For example, the EM evaluation method in the field of knowledge question answering is not applicable to the field of image recognition. The differences between evaluation methods lead to no general formula to evaluate the agent. smart

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A measure of intelligence for non-human agents
  • A measure of intelligence for non-human agents
  • A measure of intelligence for non-human agents

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] The above method regards each task / challenge in the real space as a problem, considers the expected degree of completion of each type of small problem in any problem domain, and establishes the relationship between the problem probability, the expected degree of completion and the amount of intelligence. The relationship model uses the known problem probability and expected degree of completion to obtain the intelligence of the agent. In this way, a method for measuring the intelligence of non-human agents applicable to any problem domain is provided, thereby forming a method for measuring general intelligence.

[0037] When solving the expected degree of completion of the agent, even in the same problem category, the degree of completion of the agent is also affected by various factors, such as the time and region when the problem occurs. The agent's ability prediction model with the historical completion data of the agent as independent variables can obtain the expect...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to a universal intelligent measurement method for a non-human intelligent agent, a medium, a device and a server. The method comprises the following steps: firstly, confirming anon-human intelligent agent as an intelligent measurement object and obtaining the category and probability of each problem in a target field; and then, establishing a capability prediction model of the intelligent agent for each type of problem, fitting the model through measured data to obtain current optimal model parameters, and obtaining an expected capability value of each type of problem according to the model.; and finally, substituting the problem category probability and the expected capability value of each category of problems into a general intelligent measurement calculation formula provided by the invention to obtain the intelligent quantity of the intelligent agent. Because the capability value in the method can be explained as different meanings in different fields, the method is suitable for all fields, and the method is a universal intelligent measurement method.

Description

technical field [0001] The invention relates to an intelligence measurement method for a non-human agent. Background technique [0002] With the advent of the age of intelligence, more and more independent entities are beginning to possess intelligence, they can think and interact with the environment, such entities are collectively referred to as agents in the field of artificial intelligence, and this patent divides agents into There are two categories of machines and services. [0003] However, with the rapid development of artificial intelligence, non-human intelligent bodies are showing stronger and stronger capabilities. For example, for machines, the IBM supercomputer Watson in the field of knowledge question and answering has defeated human players in quiz competitions, and Microsoft DeepImage in the field of image recognition The accuracy rate of the system in the face recognition test surpassed that of humans, and AlphaGo in the field of Go defeated Li Shishi; in ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F11/34
CPCG06F11/3447G06F11/3452
Inventor 纪雯汪宇琴
Owner INST OF COMPUTING TECH CHINESE ACAD OF SCI