Home | Videos | Contact Us   




                         << content                                      Chapter 2

Computer programs inside the conscious

The robot will sense data from the environment through its 5 senses:  sight, sound, taste, touch and smell.  Target objects are objects the robot recognizes from the environment.  The brain of the robot will extract element objects that have strong association with their respective target objects.  All element objects from all target objects will compete with one another to be activated in the mind.  These activated thoughts are known as the conscious of the robot.

There are two types of conscious thoughts:  open activation and hidden activation.  The open activations are element objects that are presented to the conscious and the robot is aware of the data.  On the other hand, hidden activations are element objects that are not presented to the conscious and the robot is unaware of the data.  Extremely complex tasks in memory, such as problem solving, will require both open and hidden activation.   

Think of the conscious as a ďmemory bubbleĒ that takes in and modify data; and the modified data instruct the robot to take action (FIG. 5A - 5B).  Inside the conscious is a computer program (or a series of interconnected computer programs) that manipulate data in the conscious.  The intelligent pathways extracted from memory and used by the robot generate this computer program.

 FIG. 5A

The intelligent pathways control the data in the conscious by adding, deleting and manipulating data.  They also control the searching and extracting of data from memory.  The instructions inside the intelligent pathways will have search functions that will extract relevant data, based on a situation, and put them into the conscious.  Some data will be extracted from memory, while other data will be extracted from long-term memory. 

There are also other things the intelligent pathways in memory can do.  They can control the comparing of data; or the analyzing of two or more data from the conscious.  These intelligent pathways can even predict the future and generate a computer program to output future events of what might happen based on the current environment. 

The conscious actually does many other things for the robot.  Some of these things include:  giving knowledge about an object, providing meaning to language, solving problems, answering questions, identifying grammar structure, following orders, planning tasks, solving interruptions of tasks, managing multiple tasks at that same time, following the rules of a game, coming up with imagination and so forth.  The intelligent pathways in memory allow the conscious of the robot to do all these things (and more). 


Important data in the conscious

The conscious does many different things for the robot.  Four of the most important things the conscious does are:  1.  manage tasks.  2.  establish rules to follow, based on the tasks.  3.  planning steps to achieve tasks.  4.  know identity.  Referring to FIG. 6, there are four containers the computer program in the conscious generated as a result of intelligent pathways:  the task container, the rules container, the planning container and the identity container.  All data from all four containers influence each other one way or another.  For example, the rules will influence what tasks to follow/abort and the planning information will influence what rules to follow or what tasks to do.

These containers are just temporary caches inside the conscious that was generated by intelligent pathways in memory.  Based on the current environment, the robot selects an optimal pathway from memory and that optimal pathway has instructions to create containers so that groups of data could be manipulated and logical thoughts and actions can be had by the robot.  The intelligent pathways create any type of computer program or discrete mathematical functions to manipulate data in the conscious -- a database system, an operating system to manage multiple threads, a word processor, an image processor, a search engine, or any software program. 


Storing and retrieving data in memory

In current database systems, a computer programmer has to create the storage functions and the search functions.  In a human robot, the brain is made in such a way that the storage and retrieval of data is based on learned knowledge.  Teachers teach the robot how to store data in memory and patterns within these lessons will establish how to store and retrieve data in memory.             

Data in memory are based on pathways Ė the robot learns knowledge by taking in movie sequences from the environment through its 5 senses:  sight, sound, taste, touch and smell.  These pathways can store static data, which are data in pathways that exists.  For example, the robot is looking at a TV guide for this week.  The TV guide for this week is one static data stored in memory.  Last weeks TV guide or the weeks before are also individual static data.  Pathways can also store linear data.  For example, the linear steps to solving the ABC block problem is linear data Ė these pathways store the linear steps the robot has to follow in order to solve the ABC block problem.

The data in memory can also be in any 5 senses:  sight, sound, taste, touch or smell; and each sense can be represented in different ways.  A music song belongs to sound and an action movie belongs to sight and sound.  All these 5 sense data can be either static data or linear data (or both).         

Organization of data is done through hierarchical association.  A network of data that have relations to each other, based on the robotís experiences, will knit themselves together.  The more two objects are encountered by the robot; and the closer these objects are from one another, the more association they will have with each other.

Although data is based on hierarchical association, the real factor that organizes them in memory is intelligent pathways.  Teachers will teach the robot how to store data in memory.  They will teach the robot which are important data and which arenít.  For example, if a teacher gives a 2 hour lecture on nanotechnology, the robot will identify the most important data from the lecture and store them in memory.  These important data will have top priority, while minor or repeated data will have low priority.  Another example would be if someone tells the robot a 2 hour story.  The robot can summarize the story into 3 sentences.  That 2 hour story can be represented by 3 sentences.  The 2 hour story can also be represented by a simple fabricated movie. 


Learning forms

Letís say that there is an object composite of a person Dave in the robotís memory.  Dave is a good friend of the robot and they have been friends for many years.  The robot will store a network of data concerning dave (which is called an object composite).  How does the object composite of dave structure the data in a hierarchical manner?  How do we know that the face of dave is more important than the leg of dave?  How do we know that a name of a person is more important than their age?  How do we know that important facts about dave should be prominent in the network? 

The answer is by lessons learned in school.  Teachers taught us, through class lessons, to always focus on a personís face to identify them.  This intelligent lesson instructs us to always look at a personís face to id them.  By focusing on a personís face, the face becomes stronger and it will be the dominant encapsulated object in a human object.  Another factor is that human beings focus on the face because it gets their attention.  The voice of a person comes from the face, so we innately focus on the face. 

The last example shows us how the encapsulated object ďfaceĒ is the dominant object in the object ďpersonĒ, but what about information about a person?  How do we know that a personís name is the most important id?  The answer is from observing different fillable forms.  Notice that in all forms or database entries, the name is the first data representing a person.  By looking at forms, we can also map out the second or third most important data about a person.  We know that age, gender, phone number and address are the most likely secondary data listed in a form. 

We can also use intelligent pathways to determine what are major or minor data regarding a person.  For example, rarely does anyone want to know a personís religion or what their petís name is.  These data are considered minor data.

Intelligent pathways learn these things innately or by lessons from teachers in how data should be organized.  Referring to FIG. 7, the diagram shows an object composite (a network) on dave.  These data have relations to each other based on association.  In data1, a form of dave is present, whereby it is a static data; which lists major information about dave such as his name, phone number, age and occupation.  In data2, a visual movie of dave is present and the face of dave is the most prominent data.  The movie sequence can be static data or linear data.  It could be an experience the robot had when encountering dave in the past.  The last data, data3, is a sound recording of dave when the robot was talking to him on the phone.  This sound recording has voice patterns that distinguish dave from anyone else.

FIG. 7 


Lessons learned in school to organize data in memory        

Lessons in school will create intelligent pathways in memory that will interpret data.  If the robot had to listen to a 2 hour lecture from a boss; and in the two hours, the boss has given 1 command sentence, the robot is intelligent and will id and remember the command sentence.  All other data in the 2 hour lecture is minor data, but the important data is the command sentence.  The robot can also use intelligent pathways to summarize and interpret what the boss wants the robot to do in the future.  

Maybe, in the 2 hour lecture, there is no clear command sentence or the boss doesnít specifically tell the robot what to do.  The boss is telling the robot ambiguously what he should do in the future.  Intelligent pathways in the robotís memory interpret the 2 hour lecture and summarize it so that the robot understands the most important information.  These activated element objects are the robotís conscious that are generated by intelligent pathways.  These conscious thoughts help to better organize data in memory.

Referring to FIG. 8, in the 2 hour lecture, the robot was able to activate conscious thoughts that relate to objects, events and actions.  Some of these activated thoughts are logical analysis, summarizing of events, additional data related to objects and so forth.  These activated thoughts (or element objects) help organize data in memory when the current pathway self-organize in memory.  Objects, events and actions are delineated and boundaries are set, whereby individual data are hierarchically structured.  For example, the 2 hour lecture is grouped as one data.  Within the 2 hour lecture are topic1, topic2 and topic3.  Within each topic are their respective inner groupings.  Additional data like summarizing and id of an important command sentence also help group data in memory in an intelligent manner. 

 FIG. 8                             Activated thoughts to organize data


Thus, when the robot wants to search for data on the 2 hour lecture experience, he can search the actual pathway45 or search the activated thoughts (element objects) to find data.  In fact, if the robotís brain searches the activated sentences, the search process will be much faster.  For example, if topic2 was a speech on nanotechnology, the search function will go into topic2; and since topic2 has reference points to the beginning and ending of topic2, it can search for more specific data on topic2.  This would be much faster than searching all data in pathway45.     

Pathways in memory forget data and pathway45 can break up into a plurality of fragmented pathways.  The activated element objects help the robotís brain to reference data even when pathway45 is forgotten.  For example, if the robot experience pathway45 10 years ago, he can still recall some images and activated thoughts.  He might not be able to remember everything, but the most important data is still stored in memory. 


Compatibility for different data types

All data in memory has same, similar or different data types.  Referring to FIG. 7, data1, data2 and data3 are all in different data types and the length of each data is different.  How will the search function find specific data or analyze data or compare different data types???  The answer is through intelligent pathways stored in the robotís memory.


<< content               next chapter >>



Home | HLAI | UAI | Books | Patents | Notes | Donation

Copyright 2006 (All rights reserved)