Mega Code Archive

 

Categories / Delphi / Functions

 

Functional analogue of the biological brain

Title: Functional analogue of the biological brain Question: I am an engineer from former USSR. From long time ego it was interesting to me to find out how we are thinking and if it is possible to build a functional analogue of it. Now I am now that it is possible, and actually are simpler then developing imitation of human performance for every different tasks. The main difference with an existing approach is in the new possibility to make a computer system subjective. That will make it capable to have mind on its own. I refuse to patent that approach because that process required a full disclosure. I believe that participating partners will keep all information confidential under pressure of necessity to avoid it uncontrolled spreading. For now I am looking for few people familiar with image of text recognition and database developing. On the first stage we will have to define the structure of the system and how many participants we will be needed. I have the program for demonstration and explanation how it is working. Please, contact me for further details. Michael Zeldich szeldich@netzero.net Answer: The passage from the physics of the brain to the corresponding facts of consciousness is unthinkable. John Tyndall (1871) Now mankind can receive the desirable during so long time reasonable partner capable to help with satisfaction of any imaginable desires. Michael Zeldich (2007) Dear frends, the following is my point of view on the possible ways for use the new, invented by me, class of devices as control unit. That approach to developing artificial reasonable subjective devices is representing the revolutionary step forward from the conventional computers. After deployment that kind of devices, thanks to inherited subjectivity, will not required programming and in conjunction with appropriate hardware will be capable to achieving spesific goals be guided by general motivation instead of algorithm. Introduction In our opinion in order to fulfill the requirements to the system capable of understanding what it is doing the new class of devices should be developed. That class of devices will differ from all the currently known devices in the computer field, because they will be subjective. Subjective nature could bring to that class of devices the ability to learn instead of to be programmed. That, in turn, raises the question of how to motivate them to perform the tasks which are necessary for the owner. It is known that at MIT, for example, researchers are developing the basic principles of building the robotic society where members of it could be motivated by programmed analog for human moral rules. We could propose the different way of motivation. Projected systems could be motivated according to desire of the motivator. The Basic concepts In our opinion, there is no reason to attempt to develop the Consciousness system. The theories of "Consciousness" have described its properties by explaining the behavior of the system. Every known explanation of its properties actually is a description of some act of the behavior 1 of the "conscious" system. We believe that it is possible to develop the subjective systems which will behave in reasonable way. However, the problem is not only in the designated terms. Our opinion is that the concept of "Consciousness", in a technical sense, isnt a valid one. It is a thought up, nonexistent phenomenon. The theories are attributing the various properties to "Consciousness", but it is not proven that it is possible to define it in a scientifically consistent way. Assumption that Consciousness is really existing phenomenon, inevitably leads to necessity, to place the central managing body inside of the brain, but that is in contradiction with the known facts. Nevertheless, attempts to create and study Consciousness systems still continue. The last, known to me attempt to design the elements of a Consciousness systems was based on micro-tubular structure theory proposed by professors Roger Penrose and Stuart Hameroff. As a result of that study, according to the authors of the theory, connections between a micro-tubular structure and Consciousness has not been established. We have opinion, that the term Reason instead of the controversial terms Cognitive, Consciousness or "Artificial Intelligence" makes more sense for a description of the subject. We can judge about the abilities of the system to be reasonable only by the phenomenon of its observable behavior. The hierarchy of subjective models which reflect its past experience determines the behavior of the system. The control system aspires to receive the greatest positive reaction on the part of itself, and other reasonable systems, if it is a member of the society, by establishing the models of behavior (1). The summary of the above is: "The system is reasonable, if it is capable to determine its own behavior, be guided by its own, subjective representation about the World known to it." There is no rationality without subjectivity! In opposition to V.I. Lenin statements that The world is given to us through senses it is possible to state: The World is given to us by our own subjective representation about it! Unique possible link between subjective representations about the real world and the real world are subjective representations about practice of interaction with it. There where such interaction is impossible subjective representations didnt have any limitations. We believe that creation of the reasonable systems could be directed by these ideas. The reasonable system cant be guided directly by any finite set of programs. Almost all that is necessary for the development of the discussed system could be derived from these statements. (1. the term behavior is describes all that is happening with the system under influence of its own internal causes. For example, speech, mind, perceiving, feeling at al. are kinds of behavior. That is solution of many problems in philosophy concerning relations between a material body and both non-material ideas and representations. The best way to implement such a system is with the help of some combination of specialized software tools and novel kinds of hardware, and it is further likely to be augmented with memories of various sorts (long-term, short-term, etc.) and with associated modules for perception and taking action, as appropriate for the end application. Given their abilities to process knowledge and to reflect on their own behavior, reasonable systems might be characterized as systems that know what they are doing. Reactive processes can be based on training, so system will activate appropriate models of behavior, or reflex after it will recognize a situation, to which it was trained. The rigid models of behavior will represent instincts built into the reasonable system. Reasonable system will be able to conduct reflective processes that operate on the base off its own representation about its own past experience. The system can be motivated to find the contradictions in its own subjective models of the World known to it. It also can look for the ways of solving these contradictions.