What is the difference between a person and a program
Neural networks, which now make up almost the entire field of artificial intelligence, can take into account much more factors than a person in making a decision, do it faster and, in most cases, more accurately. But programs only work as they are programmed or taught. They can be very complex, take into account many factors and act very differently. But still they cannot replace a person in decision-making. How is a person different from such a program? Here it is necessary to note 3 key differences, from which all others follow:
- A person has a picture of the world, which allows him, in terms of information, to supplement the picture with such data that are not prescribed in the program. In addition, the picture of the world is structurally arranged in such a way that allows us to have at least some idea of ββββeverything. Even if it is something round and glows in the sky (UFO). Usually, ontologies are built for this, but ontologies do not have such completeness, poorly take into account the ambiguity of concepts, their mutual influence, and so far are applicable only in strictly limited topics.
- A person has a logic that takes into account this picture of the world, which we call common sense or common sense. Any statement makes sense, and takes into account hidden undeclared knowledge. Despite the fact that the laws of logic are many hundreds of years old, no one still knows how the ordinary, non-mathematical, logic of reasoning functions. We basically don't know how to program even ordinary syllogisms.
- Arbitrariness. Programs are not arbitrary. This is perhaps the most difficult of all three differences. What do we call arbitrariness? The ability to build new behavior that is different from what we have done under the same circumstances before, or to build behavior in new, never seen before circumstances. That is, in essence, this is the creation on the go of a new program of behavior without trial and error, taking into account new, including internal, circumstances.
Arbitrariness is still an unexplored field for researchers. Genetic algorithms capable of generating a new program of behavior of intelligent agents are not a way out, since they do not generate a solution logically, but through βmutationsβ and the solution is found βby chanceβ during the selection of these mutations, that is, through trial and error. A person finds a solution immediately, building it logically. The person can even explain why such a solution is chosen. The genetic algorithm has no arguments.
It is known that the higher an animal is on the evolutionary ladder, the more arbitrary its behavior can be. And it is in humans that it shows the greatest arbitrariness, since a person has the ability to take into account not only external circumstances and his learned skills, but also hidden circumstances - personal motives, previously reported information, the results of actions in similar circumstances. This greatly increases the variability of human behavior, and, in my opinion, consciousness is involved in this. But more on that later.
Consciousness and arbitrariness
And what about consciousness? In the psychology of behavior, it is known that we carry out habitual actions automatically, mechanically, that is, without the participation of consciousness. This is a remarkable fact, which means that consciousness is involved in the creation of new behavior, is associated with orienting behavior. It also means that consciousness is connected exactly when it is necessary to change the habitual pattern of behavior, for example, to respond to new requests, taking into account new opportunities. Also, some scientists, for example, Dawkins or Metzinger, pointed out that consciousness is somehow connected with the presence of an image of oneself in people, that the model of the world includes the model of the subject himself. How then should the system itself look, which would have such arbitrariness? What structure to have so that it can build a new behavior to solve the problem in accordance with the new circumstances.
To do this, we must first recall and clarify some well-known facts. All animals that have a nervous system, one way or another, contain in it a model of the environment, integrated with the arsenal of their possible actions in it. That is, it is not only a model of the environment, as some scientists write, but a model of possible behavior in a given situation. And at the same time, it is a model for predicting changes in the environment in response to any actions of the animal. This is not always taken into account by cognitive scientists, although this is directly indicated by the discovery of mirror neurons in the premotor cortex, as well as studies of the activation of macaque neurons, in response to the perception of a banana, in which not only the banana area in the visual and temporal cortex is activated, but also the hands in the somatosensory cortex, because that the banana model is directly related to the hand, since the monkey is only interested in the fruit that he can take it and eat it. We simply forget that the nervous system did not appear to reflect the world of animals. They are not sophists, they just want to eat, so their model is more a model of behavior, and not a reflection of the environment.
Such a model already has a certain degree of arbitrariness, which is expressed in the variability of behavior in similar circumstances. That is, animals have a certain arsenal of possible actions that they can perform depending on the situation. These may be more complex temporal patterns (conditioned reflexes) than direct reactions to events. But it's still not completely arbitrary behavior, which allows us to train animals, but not humans.
And here there is an important circumstance that we need to take into account - the more known circumstances are encountered, the less variable the behavior is, since the brain has a solution. Conversely, the newer the circumstances, the more options for possible behavior. And the whole question is in their selection and combination. Animals do this simply by showing the whole arsenal of their possible actions, as Skinner showed in his experiments.
This is not to say that voluntary behavior is completely new, it consists of previously learned patterns of behavior. This is their recombination, initiated by new circumstances that do not completely coincide with those circumstances for which there is already a ready-made pattern. And this is precisely the point of separation of arbitrary and automatic behavior.
Modeling Randomness
The creation of a program of arbitrary behavior that can take into account new circumstances would make it possible to make a universal "program of everything" (by analogy with the "theory of everything"), at least for a certain domain of tasks.
To make their behavior more arbitrary, free? My experiments have shown that the only way out is to have a second model that models the first and can change it, that is, act not with the environment as the first, but with the first model in order to change it.
The first model responds to the circumstances of the environment. And if the pattern it activated turned out to be new, the second model is called, which is taught to look for solutions in the first model, recognizing all possible behaviors in the new environment. Let me remind you that in the new environment, more behavioral options are activated, so the question is precisely in their selection or combination. This is because, unlike familiar environments, in response to new circumstances, not one pattern of behavior is activated, but several at once.
Every time the brain encounters something new, it performs not one, but two acts - recognizing the situation in the first model and recognizing already performed or possible actions by the second model. And in this structure, many possibilities are manifested, similar to consciousness.
- This two-act structure makes it possible to take into account not only external, but also internal factors - in the second model, the results of the previous action, the distant motives of the subject, etc. can be remembered and recognized.
- Such a system can build new behavior immediately, without the long learning that is initiated by the environment according to evolutionary theory. For example, the second model has the ability to transfer solutions from some submodels of the first model to other parts of it, and many other features of the metamodel.
- A distinctive property of consciousness is the presence of knowledge about its action, or autobiographical memory, as shown in article (1). The proposed two-act structure has just such an ability - the second model can store data about the actions of the first (no model can store data about its own actions, since for this it must contain consistent models of its actions, and not the reactions of the environment).
But how exactly does the construction of new behavior take place in the two-act structure of consciousness? We don't have a brain or even a plausible model of it at our disposal. We started experimenting with verb frames as prototypes of patterns that our brains contain. A frame is a set of variants of actants of a verb to describe a situation, and a combination of frames can serve to describe complex behavior. The frames for describing situations are the frames of the first model, the frame for describing one's actions in it is the frame of the second model with verbs of personal actions. With us they are often mixed, because even one sentence is a mixture of several acts of recognition and action (speech act). And the very construction of long speech expressions is the best example of arbitrary behavior.
When the first system model recognizes a new pattern for which it does not have a programmed response, it calls the second model. The second model collects the activated frames of the first one and looks for a shorter path in the graph of connected frames, which in the best way, as it were, βclosesβ the patterns of the new situation with a combination of frames. This is a rather complicated operation and we have not yet achieved a result in this that claims to be called a βprogram of everythingβ, but the first successes are encouraging.
Experimental studies of consciousness by modeling and comparing software solutions with psychological data provide interesting material for further research and allow us to test some hypotheses that are poorly tested in experiments on people. This can be called simulation experiments. And this is only the first result in this line of research.
Bibliography
1. .
Source: habr.com