Intelligence

by | Mar 4, 2017

Merriam-Webster Definition

  1. The ability to learn or understand or to deal with new or trying situations: reason; also: the skilled use of reason.  The ability to apply knowledge to manipulate one’s environment or to think abstractly as measured by objective criteria.
  2. The ability to perform computer functions.

This Engineer’s Definition

  1. The ability to make logical choices. (Logical Intelligence)
  2. The ability to think, reason, design. and to have self-consciousness.  (Abstract Intelligence)

The two definitions listed here correspond (in reverse order) to the Webster definitions.  Logical intelligence is what is found in computer code.  Humans can design intelligent machines like computers and systems that use computers.  But humans do not have the capability to design abstract intelligent machines.  Many have argued that given time, we will learn to do this.  This engineer has his doubts. It is something we are gifted with, like migrating birds being gifted with the ability to find migration destinations, but neither we or migrating birds understand how we do it.

Intelligence is an entity unlike entities we understand from the physical sciences – we cannot write an equation using physics parameters that equate to intelligence.  We do not understand the relationship, if there is any, between a mind and physics.  Natural causes cannot make logical decisions, let alone perform abstraction.  Natural causes perform work that is defined solely by the laws of physics, not a mind.  Life, even though it follows the laws of physics, could not result from natural causes because of its embedded intelligence.  Only abstract entities that can perform intelligent work can produce machines.  Intelligence of any form has meaning only to an entity that has abstract intelligence.

Additional Clarification

For purposes of clarity and precision, there is a need to define different levels of capability and means of implementation of the term intelligence.  This is like the realm of space with levels of position, velocity and acceleration.  For the realm of intelligence, the levels would be information, logical intelligence, abstract intelligence and supreme intelligence.

Information

Specified Information is not intelligence per se, but it is a starting point for intelligent activity, like a point (position) is a starting point for considerations for distance, velocity and acceleration in the realm of space.

All matter can be defined by state variables that are information,  Matter (static) that has been manipulated by intelligent work, but is not a machine, is called a tool, and includes all human static creations such as entities we normally call tools such as a screwdriver, hammer or saw, but also with things like jewelry, art (paintings, pottery), glass, and building supplies (wood, nails). All these are creations by an abstract intelligent entity (humans).  A bird’s nest is made by a logically intelligent living entity (a bird).  All specified information comes from an intelligent entity.

Logical Intelligence

Logical Intelligence is the ability to perform logical functions, that is, to make choices (output) to achieve the desired end that dictates a logical algorithm that processes supplied information (inputs). 

Matter that has embedded logical intelligence is called a machine or a component of a machine.  Examples include mousetrap, engine, drill, watch, and computer.  Embedded logically intelligent devices require an abstract intelligent machine to actualize them, and the only example on earth is humans. Life also has embedded logical intelligence – it runs the life process going on in every living cell. Life, especially life with abstract intelligence is beyond the ability of humans, in my opinion.  This is the product of a supremely intelligent machine, with not only the ability to design, but build the cells.  This involves “hand building” the first molecular machines, which requires fingers orders of magnitude smaller than ours.

Neural Network Intelligence

As far as I know, virtually all animals have brains consisting of networks of neurons, axons and dentrites.1  Some years ago, a reseacher from the University of British Columbia who was studying neural networks by observing the operation of some sea worm brain that consisted of 30+ neurons.  It was fascinating.  The description explained pre-knowledge, learning, forgetting, and why memeory isn’t always instantenous.  The description also explained how the brain controls actions, and why these actions can change with experience. 

Engineers can mimic a neural network using analog circuits.  However, the quantity of such elements and interconnections required this approach impractical.  However, digital computers can mimic neural networks with much less hardware but at the expense of lots of computing and are called artifical neural networks, (ANN).2  This use of ANNs is on the rise because solutions to many problems do not lend themselves to linear logic, e.g., face recognition or control of traffic lights. 

The machines in the cell use appear (to me) to be either logic or analog

 

Abstract Intelligence

Abstract Intelligence is the ability to think, reason, design and to have self-consciousness.

The only example on earth is human beings, which can be considered an abstractly intelligent machine.

Creation of humans requires intelligence much greater than ours and fingers much smaller.

© 2016 Mike Van Schoiack

  1. see this UTEP description and this UNAM (National Autonomous University of Mexico) description that describes how these networks which are connected to various inputs (sensors, e.g., the rods and cones in the retina) and outputs (actuators, e.g., muscles) through nerves.
  2. Here is a nice explaination of artifical neural networks
Share This