|
 |
| It
is most suitable to see the paper of part3 at the first, then, if you are interested in
the paper, you may see part1 and part2. Because part3 is wonderful descriptions and
viewpoints all of these are that the peoples have been exploring and researching for many
years. Although without the base of part1 and part2, the man of that the abstract thinking
ability and space imagination is better, generally can know at approximation.
Part1 and part2 are the constructional descriptions and theoretical proofs, which are
dull. When you first see part1 and part2, and do not know the important discusses of part3
which are constructed from this base in advance, it is possible to abandon the reading of
the paper. |
| Thesis Description: |
A Theory of Holographic
Topology and
the Mathematical Model of Brain Thinking
Jiaxiang Bi |
The paper introduces a kind
of the mathematical model for artificial intelligence: holographic space, and introduces a
kind of the new subject: Holographic Topology.
The paper starts from the research of information structure, sets up a
concept of characteristic information, then a continuous, consistent and superior ensemble
texture ------- ¡°Holographic Space¡± is constituted by an organic combination of the
characteristic information in storage. Holographic space is a superior texture form of
information space on which an evolution on the internal structure of the space has
occurred. It disengages that primary random combination state the elements or subspaces of
which are non-correlate and independent in the old space, which whole space is integrated
into an organic unity, and a new important property is produced from the superior
hierarchy within whole.
The paper will discuss
and explore the problems about the holographic topology axiom system and the artificial
intelligence as follow:
1. To construct and discuss
an axiom system for holographic topology.
2. To discuss and analyze
the topologic correlation characteristic on the multi-stratification of the holographic
topology theory.
3. To discuss and analyze the some problems about artificial
intelligence, to introduce a kind of the coding method of which is peculiar to the nature
itself: natural coding method, and analyze the stratum structure characteristic and
property of the informations in the brain.
The paper constructs a axiomatic
system of holographic space, builds a kind of new space structure, in this space, the
element on some part can be implicated in all region of the whole space, therefor, the
some new and important properties which are different from the classical theory. The
compatibility between the elements in the holographic space, and the concordance
consistency between an element and its coordinate, are two most important properties of
the holographic space.
The paper introduces the sifting
method of the informations: the fine sifting method and the large sifting method. And to
make the focal point discussion for the blind zone problem in the large sifting method,
the paper proves: in the linear holographic space having continuous potency, if there does
not exist the singular region and the abnormal discontinuity among all component in , then
in the blind zone there are no better components. And by build one group, to discuss the
extension process of the structure of the holographic space and the generating of new
element.
In the second part, the paper
puts the conceptions of the holographic homeomorphism and the holographic isomorphism. And
discusses the clustering reaction in the holographic space, to prove that after the
clustering reaction, the subspace which is generated by reassembling is still on original
holographic space. And discusses the topological properties of the holographic space on
the kinetics, puts the problems about the dynamical equilibrium and equilibrium shifting
of the information elements in the holographic space.
The paper puts the two kinds of the recursion structures: the
general looping recursion structure and the nesting recursion structure, and proves that
the topological structure inside the holographic space is a kind of the recursion
structure.
The paper makes a total description to the holographic space,
the holographic space has three large hierarchy topology structures. First large
hierarchy: the information space hierarchy, which is possessed of the topological
structure of memory form. Second the large hierarchy: the holographic space hierarchy:
which the informations are transformed from disorder arrangement into order, to produce
the topological structure which it is possessed of holographic correlation. Third large
hierarchy: super-topology structure.
In the third part, the paper points out: that on processing
information, brain is to process the topological relation and the topological property of
the whole of an information set rather than the independent information elements. Because
the nature is unable to do that artificial ¡°encoding¡± technique, has to turn to the ¡°natural encoding system¡± which is characteristic of the nature proper. The paper
introduces a kind of the encoding method which is characteristic of the nature proper: ¡°natural encoding¡± system.
This is a kind of the process of group processing, in the life body, the informations do
be such encoding and processing. And by a flow diagram, describes simply the forming way,
the structure form and the transformation process.
The paper lists to compare the differences between the brain
and the computer system on the information processing. The paper points out that the most
important difference for the brain and computer system is that the of brain processing is
the topologic relations and the topologic graph of the informations, but theof machine
processing is the codes which are defined beforehand by the programmer.
The paper discusses elementally the feature and the functions of
artificial intelligence model, and puts that artificial intelligence is not to make only
the processing of codes, but needs to can make the processing of topologic graphs,
topologic relations and topologic properties. The essence of the paper is to explore the
problem on artificial intelligence from the topological structure of holographic space.
To various artificial intelligence models and neuron models,
the paper points out that there are three periods of the development of the machine
intelligence: the machine code period, the machine physiology period, the machine
psychology period. Only if the machine psychology period has been developed, to be able to
produce true artificial intelligence. Because thought form and thought process are of
topology property on the whole, which seem to be independent of the choice of the machine
type on the physical hierarchy. Whichever machine types to reach this object all is
possible. However the machine codes are electrical, lighted, numerical, simulative, even
the neuron impulses of the life, or however they are to be defined by man, or to be
obtained by the long evolution of the life in nature, if that their topological
correlation structure of the topological holographic system of high hierarchy which are
based on the machine code system are same, then their thought forms and thought processes
should be also equivalent on the topological relation.
The paper discusses the topological structure of the
stratification texture of intelligence system. To make the focal point analysis for the
stratification relations of intelligence analysis processing system, and put up
specifically the three large hierarchy structures of the intelligence system: the physical
hierarchy, the physiological hierarchy and the psychological hierarchy. These large
hierarchy themselves each have the different forms and topological structure, and every
large hierarchy can be divided into some sub-stratums again. The paper more completely
describes the relations between the three large hierarchies and the some sub-stratums, the
encoding method of informations, the holographic frame structure of the informations, the
functions of the virtual machine system on different levels, multi-stratification
recursion structure of informations, etc. in more detail.
|
|
