Intelligent Systems And Their Societies

Walter Fritz

Objectives

All ISs (Intelligent Systems) have a built in main objective. Many ISs can also learn how (or have written within their code the ability) to create and use sub objectives; lower level and/or temporary objectives. By reaching these sub objectives the intelligent system gets nearer its main objective.

Scope of "Objectives"
Are the objectives of an IS those of the body or those of the mind? After some thought, we can see that neither is true: they are objectives of the total IS. The mind is only the functioning of the brain and the brain is only a part of the IS; an important part to be sure, but still just a part. While it is true that the IS could not function without a brain, it is also true that it could not function without the digestive system, or without the heart, or without actuators such as arms and fingers, etc. From this we can see that the IS is actually dependent upon many or most of its parts together. This "interdependancy" is one of the hallmarks of what is currently known as a "complex system." The objectives and many other attributes of complex systems can not really be linked to any particual part of the system; only the entire system can own them.

Ranking Objectives
In its functioning and/or problem solving endevours, an IS often needs to be able to rank the importance of many objectives against its perception of its current situation. One way it can measure the relative importance of a given objective is by assigning a number between zero and one that can be generated with a formula such as:

I = w/86400

Where:

I is the importance of an objective.

w is the seconds per day that the IS is willing to spend reaching for the objective.

86400 is the total amount of seconds in a day.

We also need to emphasize that while this formula involves the time the IS is willing to spend to achieve an objective, there is no known relationship between this time and the actual time that will be used. It turns out that the actual time is most highly dependant upon environmental parameters. For example, drinking is a very important objective and most systems would be willing to spend a lot of time reaching this objective. Normally getting a drink takes almost no time at all, but if we are alone in the middle of a hot desert, for instance, the time required to get a drink of water is substantially greater!

Main Objective
In natural ISs, the random perils of their environment have resulted in the determination of their main objective as being the survival of the species, as was shown by Charles Darwin.

In artificial ISs, for instance in robots, the main objective is determined by the writer of the ISís software. In some of his books, Isaac Asimov  [An Isaac Asimov FAQ] (Exterior link) suggested and tested a triad of main objectives which he called the "The Three Laws of Robotics". They are:

1. A robot may not injure a human being, or, through inaction, allow a human being to come to harm.
2. A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

In other cases of artificial ISs, their main objective can be acting to maximize approval and minimize disapproval from a person. For instance, the system "The General Learner" (For continuos reading, like a book - do not enter here now) has such a main objective and receives approval and disapproval through key presses of an interacting person.

Generating Sub Objectives
Artificial and natural ISs can also create their own sub objectives. For instance, there are some artificial ISs whose sub objectives can be generated by the system's "body." They may sense "pain" when they run into an object, get "tired" when awake too long, or get "hungry" when they have not "eaten" for a time. (All these are usually represented by ranges of number values of particular present situation variables.) The programmer of these ISs has set their main objective as "to have positive sensations and avoid negative ones". These sensations then cause new, and temporary sub objectives according to which the IS attempts to change its present situation to one with "satisfied" sensations.

In humans and in animals, which are Natural ISs whose main objective is "to stay alive", this objective has multiple sub objectives which may shift in overall or maximum importance. For instance, "continued and sufficient breathing" is a sub objective of the main objective.  If something hinders us from breathing, we elevate this sub objective to the most important one and use all of our available time (seconds) to improve or obtain it.

Using Sub Objectives
Sub objectives are a certain situation which the IS wants to reach. It looks for response rules which are applicable to the present situation and which permits it to reach the subobjective situation.
We can also say that sub objectives are a certain situation which the IS wants to perceive as present situation. This has been explained very well by William Powers in his Perceptual Control Theory. (Exterior link)

For continuos reading, like a book - continue here.
Jump to the top of this document /Details of IS /Intelligent Systems /e-book Contents.