Artificial Thought Model

Development of computers can be roughly divided into two stages. In the first
stage, the Von Neumann architecture is applied for numerical computation,
document processing, and database management and query. All these
applications have specific algorithms, though somewhat difficult in programming.
The second stage focuses on symbolic and logical processing, in which
knowledge and information processing mainly bases on reasoning. How to
choose effective algorithm is the key problem to this stage of research. All these
applications are well defined and explicitly represented problems of the ideal
world. However, many real-world problems are ill-structured, such as pattern
recognition, problem solving and learning from incomplete information, etc.
These problems are in the category of intuitive information processing.
For intuitive information processing, theories and technologies of flexible
information processing should be studied. Flexibility in real world has the
following characteristics:

  • Integrate varieties of complex and intricately related information containing
    ambiguity or uncertainty information;
  • Actively acquire necessary information and knowledge, and learn general
    knowledge inductively from examples;
  • Automatically adapt to users and changing environment;
  • Self-organization based on the object for processing;
  • Error tolerant information processing.

Actually, human neural networks capable of large-scale parallel and
distributed information processing inherently support flexible information
processing. Thus, we proposed the artificial thought model in Fig. 1.3.
The artificial thought model in Fig. 1.3 clearly illustrates that artificial
thought bases on open autonomous systems, takes fully advantages of varieties of
information processing patterns to achieve collective intelligence, then proceeds
with flexible information processing, and finally solves real-world problems.

  • Actually, human neural networks capable of large-scale parallel and
    distributed information processing inherently support flexible information
    processing. Thus, we proposed the artificial thought model in Fig. 1.3.
    The artificial thought model in Fig. 1.3 clearly illustrates that artificial
    thought bases on open autonomous systems, takes fully advantages of varieties of
    information processing patterns to achieve collective intelligence, then proceeds
    with flexible information processing, and finally solves real-world problems.
  • Leave a Reply

    Your email address will not be published. Required fields are marked *