Cognitive AI~WHEEL logoPhysics
5) Entity Task Universe is the target environment, and the physical and cognitive machinery of the entity itself (including the I/O interface).


Following is reprinted from portions of: The Cognitive Ameba: Evolving Reactionary Behavior From a Single Modular Unit - Proceedings of the Second Online Workshop on Evolutionary Computation (WEC2),pp.1-4,1996

Increasing ETU Mass (getting smarter)
Effective reactionary cognitive structure (intelligence) is generated as a result of the environmental interaction of a single initial processing module composed of two basic decision generating machines. One of these primary decision machines is applied to the target environment and the other is applied to cloning, mutating, and testing clones of the basic decision entity (LDM) at various cognitive and physical observation posts (OSPs) within the three realms of this model of the entity's task universe.

Where Exactly Do These Perspective Expansions Occur?
Expansions take place, for any given LDM, within the three realms of matter and motion of the ETU. All or part of each of these three realms must be approximately defined and interactively modeled within the cognitive space of the entity.

OSPs
Creating Revealing "Observer/Shaper Posts" (OSP)
Assign new decision circuits (sub-entities) to observe, discover, and perfect value enhancing interactive order from various observation posts on (and within) the entire three part ETU model.

Evolving LDMs "tailor their processing and formatting" simultaneously within each currently existing "post on order".

All of the portions of the entity, and its task universe, are represented by the three ETU realms of action and material. Intelligence is about discovering the inherent interactive order that exists within, and between, these three portions. Some such order may be revealed exclusively when focusing on a given certain configuration, perspective, and perception.

Therefore, finding direct solutions to certain real problems may involve one or more additional observation posts on the matter and motion of the ETU. Any or all of an entity's three portions of matter and motion may be involved in configuring a new Observer/Shaper Post (OSP) expansion event. (Notice that an OSP may be observing one portion of ETU m&m, and shaping another portion. Such specifics are designated in the configuration of the given observation post.)

There is more than one way to "watch" a machine operate. One may consider all of the machine and its material, or merely part. This selected configuration of interest may also be "observed" from a number of different value inhibited perspectives.

Entity Task Universe
ETU = E + I + V
Entity + i/o Interface + enVironment

Modeling the ETU: The 3 Realms of ETU Matter & Motion
I. Entity (processor)
..A. Physical
....1. Computer hardware
....2. Portions of the i/o interface (and tool extensions)
..B. Cognitive
....1. Processing
....2. Records (including start up knowledge)
....3. RAModel of the ETU
II. Input/Output Interface (corresponding physical & cognitive portions)
..A. Sensors
..B. Shapers
..C. Energy absorption renewing conversion
(optional nutrition or energy harvesting mechanism)
III. Environment
..A. Laws and Boundaries
..B. Elements (including nutrient or renewing specifics)
..C. Artifacts (tools or extension of the i/o interfaces)
..D. (other entities)

Creator assigned "value or objective" might easily be argued as a fourth element the ETU. Value denotes the polarity of event records (cognitive structure) from the first flux cycles of the first LDM, and influences the direction of expansion by qualifying the more "effective" observation posts. Objective might rationally be viewed as a portion of the ETU. I have not included entity values because this appears to be a constant "directional" force, as opposed to a realm of m&m. (Notice that the entity's gauges on this prevailing constant may be mutable.)

The creator assigned definition of value is the "magnetic north" that shapes all three realms of the ETU as a product of their interaction. Of course assigned objectives may differ from one Observer/Shaper Post to another, but value will normally be finally assessed in the target universe of the zero module (the most senior and nucleic LDM module that defined the dominant ETU).

There are different ways to define this constant of "value and/or objective". The creator may simply supply the entity with a direct "compass". A game entity (for instance) would evaluate high score or victory as direct indicators of value; or value may be creator designated as survival, continuity, and/or propagation of the entity. Value may be vaguely, abstractly, incompletely, or indirectly defined; but some indication of value must be provided by the creator. More precise definitions of value will of course provide for more directly discovered and effective cognitive reactionary structures.

IX. EVALUATING THE EFFECTIVENESS OF A GIVEN ISOLATED SELF-IMPLEMENTED SHAPING EVENT
(output action sequences)

Reviewing of the Effect of Reflected Order,
Including Resected (Created) Order
A. Feedback Perceptions, of the Effectivity of Self-Produced Order

MEASURING SELF-INDUCED FLUX
The effect on modular status is observed, recorded, and evaluated as a theoretically self-induced fluctuation in modular status.

Feedback Perceptions Of The Effectivity On Modular Reality Of Self-Produced Solutions

Monitoring the effect of the implementation of an intra-modularly generated solution on the local modular universe is the aspect of "self-awareness" that completes and closes the polarized processing circuit.


Copyright © by David Albert Harrell