Authors: Scott Lifan Gu
Do computers already have human level intelligence? Could they understand and process the semantics of irrational numbers without knowing the exact values ? Human can. How about uncountable sets ? These are necessary to build sciences and real world modeling. Does human intelligence exceed the power of Turing Machine? This paper explains that behavior-based Turing Test cannot measure some intrinsic human intelligence, due to the bottleneck in expression, the bottleneck in capacity, and black box issue, etc. And it does not provide a progressive measurement up to human level intelligence. Similar issues exist in other current testing methods, due to the limitations of behavior-based, knowledge-based or task-based, etc. Measurements based on intrinsic mechanisms could provide better testing. This paper identifies several design goals, to further improve the measurement. Gu Test, a progressive generic intelligence measurement with levels and potential structures, is proposed based on these goals, to measure the intrinsic mechanism for semantics, potential and other intelligence. The semantics of irrational numbers and uncountable sets are identified as two test levels. More work need be done to expand the test feature sets and structures, and provide some suggestions for the direction of future Artificial Intelligence (AI) researches.
Comments: 7 Pages. The earliest version of this paper is at: https://ia600801.us.archive.org/29/items/GuTest/GuTest.txt
[v1] 2012-11-17 20:07:28
Unique-IP document downloads: 137 times
Add your own feedback and questions here:
You are equally welcome to be positive or negative about any paper but please be polite. If you are being critical you must mention at least one specific error, otherwise your comment will be deleted as unhelpful.