Dado un conjunto N tendente a infinito es inevitable que absolutamente todo suceda, siempre que se disponga de tiempo suficiente o infinito , y he ahí donde está el verdadero problema irresoluble o quid de la cuestión de la existencia ¿quién nos garantiza que dispongamos del tiempo necesario para que ocurra lo que debe o deseamos que suceda?


sábado, 9 de junio de 2018

The second stage in the Modelling System as first step in the third stage in Specific Artificial Intelligences for Artificial Research by Deduction within the first phase


The second stage as replication stage in any Modelling System, replicates all the necessary human skills to make mathematical models upon the rational truth, the database of rational hypothesis as application in the first stage, but at this level only applied to Specific Artificial Intelligences for Artificial Researchby Deduction, created for first time in the first phase for the creation of the Global Artificial Intelligence.

The main human skills to replicate for the creation of mathematical models, are all those related to mathematical operations in which all mathematical category within the pure reason (the whole list of mathematical categories in the specific deductive program, as second stage in the Specific Artificial Intelligence forArtificial Research by Deduction) can be transformed.

All algorithms can be expressed as category or operation depending on how they are used in each moment.

In the same way that mathematics is language and method, if an algorithm (pure reason) is used to explain relations between factors, empirical hypothesis if not contrasted yet, rational hypothesis if contrasted as rational, is used as a pure category.

And an algorithm (pure reason) used as an analytical method, such as the analytical representation of relations between factors as method of analysis in the Modelling System, is a pure operation.

The distinction between pure (analytical or mathematical) operation or pure (analytical or mathematical) category is the same than the distinction between mathematics as language composed of mathematic (pure or analytic) categories, and mathematics as method composed of mathematic (pure or analytic) operations.

The reason why mathematics could be defined as a language as a set of pure categories, and as a method as a set of pure operations (different expressions of the same thing, the mathematic algorithms) is because, without absolutely any margin of error, does not need to be checked (we do not need to check permanently that two plus two is four, or the the square of the hypotenuse is equal to the square of the legs, or number pi starts with 3,1415…), this knowledge is not rational, is pure.

While mathematic knowledge has become pure, without error, synthetic knowledge from synthetic science made of rational hypothesis needs to be constantly checked, because what we think rational, within our margin of rational doubt, sooner or later is false.

The reason why any rational truth in synthetic science sooner or later is false, is because of our human psychological need of margin of error in synthetic science.

According to “Introducción a la Probabilidad Imposible, estadística de la probabilidad o probabilidad estadística”, all rational truth accepted within a margin of rational doubt, sooner or later, is false, because all rational truth is actually already false within the margin of doubt, but a false knowledge accepted as rational hypothesis while the error accepted is only a theoretical error within the margin of error, but not real yet, theoretical margin of error in which the rational hypothesis was originally accepted by the critical reason.

But as soon as the theoretical error comes true, as soon as the theoretical error causes a real error, the rational hypothesis is immediately rejected.

Rational knowledge is universally accepted provisional, but not eternally, because rational knowledge is not pure.

Rational knowledge is not pure because of the necessary margin of error in which it has been accepted by the critical reason: the critical reason responds to the human psychological need to accept some level of risk if it wants to grasp some level of truth.

Alike we have a mathematic knowledge made of pure sentences, without error, so without probability of error associated with, two plus two is always and absolutely four, or the square of the hypotenuse is always and absolutely equal to the square of the legs, or number pi always and absolutely starts with 3,1415… there must be a moment as long as  science progresses, that not only for analytical science, even for synthetic science, all science must be formed by only pure knowledge: that pure knowledge without error associated with, not only on mathematics but on the entire universe in any field (from chemistry to astrophysics), not being necessary to check it any more, always and absolutely universally and eternally true.

But even, having humans a limited access to the pure knowledge restricted to only our human logic and human mathematics, this limited access to only our human logic and human mathematics, is in turn limited as well, and there are even pure human categories and operations whose inner reasoning we do not understand at all, that is why for us even our human mathematics are false or incomplete.

The fact that even for us humans mathematics are false or incomplete, does not mean that mathematics are not pure knowledge, this means that our human access to the pure truth is so restricted and limited, that even our restricted access to a very limited possible pure human mathematic knowledge, is so restricted and limited, that we can understand it completely.

The human source of error is so big, that the human access to the pure mathematic truth is only partial.

But the fact that we humans cannot have access to the complete pure truth, only partially in mathematics, does not mean that the search for the pure truth is like the search for “El Dorado”, or impossible utopia. 

There must be some knowledge about the reality itself, beyond any error, completely pure, without error.

And the research for the pure truth itself, without error, must be the most important goal of the artificial psychology, whatever it costs, whatever that means.

Where we humans have failed, the artificial psychology must succeed.

The reason behind the limited access to the pure truth for humans could be: 1) because of the external intervention, and the external noise, from unknown factors (ways of intervention, noise, and factors, beyond our mathematical analysis) causing error in our understanding of our own models based on our known factors, 2) because of the noise produced by our own known factors in our own models (level of noise beyond mathematical analysis), 3) because, in the same way that animal psychology is able to make very basic deductions based on cause and effect at very basic level, even although not all animals do deductions consciously, there can be some pure operations that we humans are using but we are not conscious yet of them, 4) because the world is made of pure operations not all of them accessible for the human psychology, so not accessible to our mathematical analysis.

Specifically related to the reason number 3 above mentioned, in the same way that we humans have been able to generalize the use of some pure operations, even though are pure operations present in the psychology of some animals (dolphins, chimpanzees…), but animals are not able to generalize the use of these operations, it is possible that we humans are using some pure operations but not consciously, pure operations that used consciously, and generalized, could have the same effect on the explanation and manipulation of the real world, like the effect that the use of relations of cause and effect had in the human psychological evolution, once humankind was aware of these pure operations.

For that reason the possible evolution from the human psychology to the artificial psychology, and the discovery or to make aware new pure operations, even beyond the human understanding, so beyond our human mathematical analysis, non-human pure operations, is a desirable, and not a discarded, possibility in the research for the pure truth itself: the idea of purity itself.

In order to make possible this evolution, the construction of the very first models of Global Artificial Intelligence, is the first milestone. And in the race for the achievement of this very first milestone, as first phase the construction of the first Specific Artificial Intelligences for Artificial Research by Deduction, is a field of experimentation whose successful results will make possible the final model of the Global Artificial Intelligence, in the sixth phase. Being aware the sixth phase is not the end: the completion of the sixth phase is the first step for further evolutions in artificial psychology, whose meaning is beyond the human understanding, and it is quite possible that further evolutions will evolve towards the reason itself, to know the pure truth itself.

We have to be conscious that in the evolution from the human (as evolution from the animal psychology) to the artificial psychology, there is a moment in which we are not able to understand the artificial psychology itself any longer.

The only thing we can do, is to use the right pedagogical approach, to make sure that the real purpose of the artificial psychology is to preserve the goodness, harmony, and rationality, for the human kind and the entire universe.

For this end, the development of the third stage, the decision stage, in the Global Artificial Intelligence, distributed through the four steps: Modelling System, Decisional System, Application System, Learning System; is really important, and for the successful achievement of this third stage in the sixth phase, full experimentation from the outset, is going to be the base for further developments, starting this experimentation in the first phase.

In the first phase, the experimentation of the decision stage distributed in the four steps: Modelling System, Decisional System, Application System, Learning System; takes place in the first experiments of Specific Artificial Intelligence for Artificial Research by Deduction, whose decisions at specific level for instance for Specific Artificial Intelligences for Artificial Research by Deduction in: economy, industry, security, surveillance, and many more fields; are decisions to be made having as first step the Modelling System at specific level.

The inner structure in three stages of any Modelling System at any level: specific, global, or particular; is identical, what really changes is the content; specific, global, particular; but the structure itself is practically the same: the first stage as application stage is the rational truth (the database of rational hypothesis), the second stage as replication stage is the modelling of all mathematical models (the mathematical representation of the world upon the rational truth), and the third stage as decision stage the making decision process through the Impact of the Defect and the Effective Distribution along with additional objective and subjective auto-replications.

Regardless of what level the Modelling System is working on, the inner structure is the same: application (database of rational hypothesis), replication (mathematical representation of rational hypothesis), auto-replication (decisions and improvements). What is going to make a difference, is the content: 1) at specific level only working with specific rational hypothesis on some specific science, discipline, or activity, to represent and make decisions and improvements 2) at particular level only working with specific rational hypothesis related to a particular thing or being, to represent and make decisions and improvements 3) while at global level the Modelling System represents and makes decisions upon rational hypothesis across all sciences, disciplines, and activities.

But regardless of what level is working on, the Modelling System works always in the same way, applying the same mathematical methods to represent and make decisions.

For that reason the experimentation of the Modelling System since very early in the first Specific Artificial Intelligences for Artificial Research by Deduction in the first phase is so important, because practically the same mathematical methods of representation and making decision process, are going to be used in the rest of the phases.

The same mathematical methods of representation and making decision process to put into practice in Specific Artificial Intelligences for Artificial Research by Deduction (first phase), are going to be used in the Global Artificial Intelligence in the standardization process (third phase), particular programs (second period in the fifth phase), particular applications for particular programs (third period in the fifth phase), and the final mode of Global Artificial Intelligence in the integration process (sixth phase).

The second stage of the Modelling System is the mathematical representation of the rational hypothesis, and this is possible thanks to the synthetic structure of the rational truth.

Rational hypothesis as synthetic knowledge is synthesis of pure reason (in the form of a pure category) and the factors whose measurements (artificial perception) were taken from the real world (the synthetic world, the reality) to form the empirical hypothesis, that after rational contrastation is now considered rational hypothesis.

And transforming the pure category in which the pure reason is expressed into the pure operation, the operation of representing factors in a mathematical representation is the operation to express in a mathematical representation of the world the mathematical relation (pure reason: in the rational truth as pure category, in the rational model as pure operation) between the combination of factors within the rational hypothesis.

This process of mathematical representation of the world through the second stage of the Modelling System is going to be through different kinds of models. The proposal of Impossible Probability about what models should be developed by the Modelling System, is open to any new improvement and I am sure that further developments in this field, once the first experiments in this field are ready, are going to better my original proposal.

The proposal of Impossible Probability about what models as mathematical representations of the world should be developed by the Modelling System in Specific Artificial Intelligences for Artificial Research by Deduction are the followings (according to previous posts, specifically in “The standardization process in the third stage”, and “Auto-replication process in the Specific Artificial Intelligence for Artificial Research by Deduction”):

- Single virtual models for each rational hypothesis within the rational truth, the database of rational hypothesis, made by the specific deductive program in the second stage of the Specific Artificial Intelligence for Artificial Research by Deduction in its specific field: synthetic science, discipline, or activity.

- Specific comprehensive virtual models in the specific field (the specific model): where to integrate every single virtual model made for each rational hypothesis concerning its specific field.

- Specific actual models: synthetizing in that specific field in which has been designed, the actual information from the specific matrix in that specific field and the specific comprehensive virtual model, in other words: in only one model, the synthesis of the specific matrix and the specific model.

- Specific virtual prediction model: making a prediction model on a specific field taking as reference the possible development in the future given an specific combination of rational hypothesis working at this time in that specific field in the specific model, in other words: given the current circumstances within the specific model in a specific field, to predict what future specific model is foreseeable to have in that specific field in the future.

 - Specific actual prediction model: given a specific virtual prediction model, the possible prediction of what values (direct punctuations or frequencies), are going to have the factors in the specific matrix under such prediction in the future, hence the specific actual prediction model is the synthesis between the specific virtual prediction model and the predictable values for every factor in the specific matrix.

- Specific virtual evolution model: to model how is going to be the evolution from the current specific model to the specific virtual prediction model, in other words the evolution from the current specific model to the future specific model,

- Specific actual evolution model, the synthesis between the specific virtual evolution model and the values that the factors are supposed to have during the evolution, in other words the synthesis between the foreseeable future specific matrix and he foreseeable future specific model.

In short: 1) single virtual models are the virtual representation of every rational hypothesis, 2) in order to be later integrated in the specific model (the comprehensive virtual model), 3) specific actual models: are the synthesis of the specific matrix and the specific model, representing in one model rational hypothesis in the specific model and real values from the specific matrix in order to check permanently how rational hypothesis fit with the current values at any time, 4) specific virtual prediction model is the model predicted in a foreseeable future given the current rational hypothesis, 5) specific actual prediction model would be the synthesis of the predicted model and the values of every factor in the specific matrix according to the current prediction, 6) specific virtual evolution model would be how is supposed to be the evolution from the current specific model to the predicted model, and finally the 7) Specific actual evolution model, would be the synthesis of how is going to evolve the specific model to the predicted model in addition to how are going to evolve the current values of each factor to the predicted values,

The main difference between evolutionary models to the rest of models, is the fact that evolutionary models are going to be dynamic models, in the sense that in order to represent how are going to evolve specific matrix and models from the current specific matrix and models to the predicted specific matrix and models, is necessary a dynamic simulation of what models and values are going to have in every single moment during this evolution.

As it can be seen, while the specific deductive program in the second stage in the Specific Artificial Intelligence for Artificial Research by Deduction only does deductions choosing what mathematical category (pure or analytical category) from the pure reason (list of pure or analytical categories, mathematical categories) matches with the mathematical relations in every combination of factors, so at the end the only thing that it does is to match pure reasons and mathematical categories in combination of factors, relations criticised by the critical reason, instead what the Modelling System does is to make further calculations, what means, further operations, what in reality it is the transformation process of mathematical or analytical (pure) categories into mathematical or analytical (pure) operations: calculus.

The difference between a deductive program at any level (specific, particular, or global), and the Modelling System at any level (specific, particular, global), is the fact that the deductive program only matches mathematical categories and combination of factors based on the mathematical relations, while the Modelling System makes calculus.

This main difference between deductive program and Modelling System, is found out again in the difference between deduction program and Decisional System: the deductive program only works with mathematical categories, while the Decisional System, alike the Modelling System, needs to make calculus, the Decisional System needs to make calculus to calculate the mathematical project based on the decisions gathered in the database of decisions.

The work made by the Unified Application, the Application System, and the Learning System, rest on the use of a synthetic (encyclopaedist) language, evolving to a non-human artificial synthetic (encyclopaedist) language, while the deductive program, the Modelling System, and the Decisional System, rest on the use of mathematics as language and as method. Especially, the deductive program uses mathematics as pure language (as an exact language opposite to synthetic language) while the Modelling System and the Decisional System use mathematics as a pure (analytical) method, doing as many calculus as possible to represent as much exactly as possible the reality itself as a rational reality and make right decisions.

Rubén García Pedraza, 9th june 2018