This paper describes a detailed cognitive structure and related processes that it might use. It also suggests some specific methods for carrying out those processes, where the methods interact well with each other. The main purpose of this paper is to reaffirm earlier research on knowledge-based 'concept trees' and experience-based 'symbolic neural networks' by showing how they can be used as part of the described system in more detail. While an earlier paper defined 3 levels of functional requirement, this paper defines physical and conceptual levels, based mostly on reinforcement and basic counting mechanisms. The counting mechanism is used slightly differently, to measure a level of 'cohesion' instead of a 'correct' classification, over whole pattern instances. The structure is therefore modelled closely on the human brain, where the pattern creation and activation can also be automatic and the information can flow seamlessly between the two main constructs.
from cs.AI updates on arXiv.org http://ift.tt/1LlCiaY
via IFTTT
No comments:
Post a Comment