"As I have evolved, so has my understanding” from I robot.
This blog reflects my basic understanding in artificial intelligence. Since beginning many people trying to solve the riddle of life (intelligence) but none had succeeded till now. I am also one among them. Hence this field is uncertain till this riddle is solved. This makes things convenient to me since I can think independently without having any constrain.
Introduction
In this blog I am going to talk about the philosophy of artificial intelligence i.e. Purpose or motivation of artificial intelligence and good & bad outcomes of artificial intelligence.
People follows different approaches in there thinking I can’t say which approach is right to follow but I think the combination of all the approaches will definitely lead to artificial intelligence.
Hence it needs all-round understanding in all the approach which I am mentioning. So my advice to research community is this. They should work together to obtain better results.
Programming approach.
Learning approach.
Evolution approach.
Replication of human brain (neural network).
Cybernetic approach.
Swarm approach (swarm intelligence).
Genetic approach.
Bionic approach.
Ecology approach.
Electronic approach.
Philosophical approach.
In this blog I’m going to illustrate it in all the above aspects. And some of the ideas can be repeated or interlinked but there approach is different.
Programming approach
This is the oldest approach. Programming people used to give lots of instruction (syntax) or set of rules to a computer or a system to solve the problems. It’s quit familiar.
Learning approach
This is one of the new approaches in dealing with artificial intelligence. There are three methods by which a computer or a system can learn.
Learning by seeing the environment.
Learning by committing mistakes and find a way to rectify it.
Learning by creating an virtual environment (simulation)
The beauty of Artificial Intelligence programming is that the programmer cannot predict the behavior of the program proceeding with time. Since the program evolve by itself as time pass by in analogy with human brain (from an embryo to a developed man).
Evolution approach
This is also one of the new approaches in dealing with artificial intelligence. In this approach we have to think in terms of evolution of living organisms.
Mathematician John Conway’s “Game of Life” (Gardner, 1970). The Game of Life is a grid of binary elements, maybe checkerboard squares or pixels on a screen, arranged on a two-dimensional plane, with a simple set of rules to define the state of each element or “cell.” Eac
h cell is conceptualized as belonging to a neighborhood comprising its eight immediate neighbors (above, below, to the sides, and diagonal). The rules say that
If an occupied cell has fewer than two neighbors in the “on” state (which we can call “alive”), then that cell will die of loneliness—it will be in the “off” or “dead” state in the next turn.
If it has more than three neighbors in the on state, then it will die of overcrowding.
If the cell is unoccupied and it has exactly three alive n eighbors, it will be born in the next iteration.
The Game of Life can be programmed into a computer, where the rules can be run iteratively at a high speed (unlike in Conway’s time, the 1960s, when the game had to be seen on a checkerboard or Go board, or other matrix, in slow motion). The effect on the screen is mesmerizing, always changing and never repeating, and the program demonstrates many of the features that have come to be considered aspects of artificial life.
The ubiquitous “glider” pattern in the Game of Life. (cellular automaton)
Replication approach
This is also one of the new approaches in dealing with artificial intelligence. In which the neuron cells present in human brains are replicated or its properties are replicated (known as neural network & fuzzy logic).
The basic neuron consists of synapses, the soma, the axon and dendrites. Synapses are connections between neurons - they are not physical connections, but miniscule gaps that allow electric signals to jump across from neuron to neuron.
These electrical signals are then passed across to the soma which performs some operation and sends out its own electrical signal to the axon.
The axon then distributes this signal to dendrites. Dendrites carry the signals out to the various synapses, and the cycle repeats.
A schematic diagram of an ideal biological neuron. Each neuron receives multiple inputs through its dendrites and generates a single output along its axon.
A schematic diagram of five interconnected neurons. Each neuron receives numerous parallel input signals through its dendrites and yields a single output that is transmitted to other neurons.
In fuzzy logic the processing is parallel unlike our exiting computer system which produces random signals example:
Is this a circle? Yes and no. In some ways it is a circle and in other ways it is not. It is a fuzzy circle.
It is also sometimes alleged that fuzziness is another word for probability, but it is an entirely different form of uncertainty. For instance, take a second to draw a circle on a sheet of paper. Now look at it. Is it really a circle? If you measured, would you find that every point on the edge was exactly the same distance from the center? No-even if you used a compass you would find that variations in the thickness of the line result in variations in the radius. So your circle is not really a circle, it is sort of a circle (see Figure). If you did really well, it may be 99 percent a circle, and if you drew a square or a figure eight or something un-circle like, it might be closer to 1 percent a circle maybe it has rounded sides or encloses an area. It is not meaningful to say that it is “probably” a circle (unless you mean that you probably meant to draw a circle, which is an entirely different question). Fuzzy logic has turned out to be a very powerful way to control complicated processes in machines and to solve hard problems of many kinds.
In my point of view I think. When we draw a circle we are not focused on how we are going to proceed to the next point (as computers do with the help of mathematical formulas). Instead we always focus on whole circle.
Cybernetic approach
The field of science concerned with processes of communication and control (especially the comparison of these processes in biological and artificial systems). So cybernetics acts like a bridge that links artificial machines with living organisms like human. Hence control and communication with artificial intelligence may be possible.
Swarm approach
Swarm researchers are especially interested in the interactions between individual- and group-level phenomena; for instance, a bird flock has properties over and above the properties of the birds themselves, though there is of course a direct relation between the two levels of phenomena.
I can illustrate this phenomenon in other aspects also.
Each individual living cells work as a whole unit in order to survive (in the case of multiple cell organisms).
Each individual living organisms work in a group in order to survive (as in a case of ants human society etc).
Some tropical termites are able to build elaborate domed structures that are begun as pillars; in the course of building, the pillars are tilted toward one another until their tops touch and they form an arch. Connecting arches results in the typical dome. As it is frequently remarked that the invention of the arch was a major milestone in the development of the architecture of civilized man, we might wonder how in the world a swarm of simple-minded termites could accomplish the feat.
If we were building an arch, we would start with a plan, that is, a central representation of the goal and the steps leading to it. Then, as the work would probably require more than one person (unless it was a very small arch), a team of workers would be organized, with the architect or someone who understands the plan supervising the laborers, telling them where to put materials, controlling the timing of the ascension of the two pillars and their meeting. We are so dependent on centralized control of complex functions that it is sometimes impossible for us to understand how the same task could be accomplished by a distributed, non centralized system.
It appears that the termites build a dome by taking some dirt in their mouths, moistening it, and following these rules:
Move in the direction of the strongest pheromone concentration.
Deposit what you are carrying where the smell is strongest.
After some random movements searching for a relatively strong pheromone field, the termites will have started a number of small pillars. The pillars signify places where a greater number of termites have recently passed, and thus the pheromone concentration is high there. The pheromone dissipates with time, so in order for it to accumulate, the number of termites must exceed some threshold; they must leave pheromones faster than the chemicals evaporate. This prevents the formation of a great number of pillars, or of a wasteland strewn with little mouthfuls of dirt.
The ascension of the pillars results from an autocatalytic or positive feedback cycle. The greater the number of termites depositing their mouthfuls in a place, the more attractive it is to other termites. Autocatalysis is a significant aspect of many complex systems, enabling the amplification of apparently trivial effects into significant ones. As termite pillars ascend and the termites become increasingly involved in depositing their loads, the pheromone concentration near the pillars increases. A termite approaching the area then detects the pheromone, and as there are multiple pillars and the termite is steering toward the highest concentration, it is likely to end up in the area between two pillars.
It is attracted toward both, and eventually chooses one or the other. As it is approaching the pillar from the region in between, it is more likely to climb up the side of the pillar that faces the other one. As a consequence, deposits tend to be on the inner face of the pillars, and as each builds up with more substance on the facing side, the higher it goes the more it leans toward the other. The result is an arch. Termite builders are one kind of self-organizing system.
There is no central control, the intention of the population is distributed throughout its membership—and the members themselves are unaware of the “plan” they are carrying out.
Actors in the system follow simple rules, and improbable structures emerge from lower-level activities, ** (I disagree this idea because if you disturb ant’s colony I will not follow this rule as a dummy instead it will attack us)
Pheromones accumulate on the shorter path because any ant that sets out on that path returns sooner.
Genetic approach
The branch of biology that studies heredity and variation in organisms is known as genetics. By this approach the living organisms code (a program which is use to run or animate) which is embedded in its DNA can be interpreted properly so that it can be implemented in artificial intelligence in near future.
In my point of view DNA is substance (polymeric chain) which contain all the information (over all sketch) about that living organism and it act as a catalyst to build the living organism as encoded in it. Rest of the job is accomplished by the cells which follow the coded paten of DNA to build the organism. The coded paten may be modified in the DNA slightly according to the feed back got by living organism (based upon the advantages & disadvantages of that living organism’s structural design) and this can be felt in the succeeding generation of that living organism both in its character and its structure.
Molecular structure of the DNA double helix in the standard (A) A space-filling model, in which each atom is depicted as a sphere. (B) A diagram highlighting the helical strands around the outside of the molecule and the AT and GC base pairs inside.
Replication of DNA. (A) Replication of a DNA duplex as originally envisioned by Watson and Crick. As the parental strands separate, each parental strand serves as a template for the formation of a new daughter strand by means of AT and GC base pairing. (B) Greater detail showing how each of the parental strands serves as a template for the production of a complementary daughter strand, which grows in length by the successive addition of single nucleotides to the 3' end.
Bionic approach
Application of biological principals to the study and design of engineering systems (especially electronic systems).
This can be used to replace damaged parts of human (or a machine) and efficient electronic machineries can be designed.
The difference between cybernetics & bionics
Confusion between “cybernetics” and “bionics” arises from the fact that bionics are advanced cybernetic systems - bio-mechanical devices that connect to and/or replace living body parts with machine parts to make the individual more powerful, faster, stronger, etc. Like medicinal and commercial cybernetics, these device are design to respond to neurological impulses and function like an actual human organ or limb. The distinction, however, is that bionics has come to mean those bio-mechanical devices that significantly augment/enhance performance about and beyond the “normal” or “average” human spectrum.
Ecology approach
The branch of biology concerned with the relations between organisms and their environment. This study may be used to know how a living organism behaves with change in environmental conditions and how it maintains equilibrium.
We can think in terms in other dimension also i.e.
It is also possible that the planet earth is a living thing which raised living species (according to us) to maintain its body (i.e. atmosphere) in an equilibrium state (to maintain constant temperature so that it does not get disintegrated in course of time) in analogy with micro organisms living in our body to maintain our health condition stable.
Electronic approach
As we all know electronics paid the way for recent advancement in science and technology but this is not enough to develop an efficient artificial intelligence. Hence researches are going on to develop very small, compact and energy efficient electronic devices in the limelight of nanotechnology.
Philosophical approach
In this approach the artificial intelligence is treated in a philosophical manner
Philosophy of Artificial Intelligence
Imagine. How it will be, after human beings create artificial intelligence. “Does he will feel like god? Or does he believe in god there after?”
Since the time of human thinking, philosophers played a vital role in framing rules that had minimize the chaos among people. In fact they are the fore fathers of science. They are the people how think in unusual manner which gives rise to creativity or new invention. In the same way we also need philosophical thoughts when we are dealing with artificial intelligence.
Some of the philosophical questions are
What is Life?
What is intelligence?
How can we differentiate a living thing with a nonliving thing?
What is a consciousness?
What will be an objective of artificial intelligence?
What is an emotion?
What is learning?
What is love?
How can we motivate artificial intelligence?
Is there any perfect laws by which artificial intelligence may be leaded?
What is randomness (if we flip a coin in order to introduce randomness into a decision making process. If the direction and magnitude of the force of the thumb against the coin were known, as well as the mass of the coin and the distribution of density through its volume, relevant atmospheric characteristics, and so on, the trajectory of the coin could be perfectly predicted)?
How system (our environment) maintain equilibrium without any central force (even if any organism in an ecological environment might be uprooted or killed. Whole species may become extinct. Volcanoes, hurricanes, stock market crashes may result in devastation, but the system repairs itself and returns to a stable state, without central control)?
Why is there something rather than nothing?
Might the world be an illusion or dream?
What exists beyond the human senses?
What happens after death?
Does divine or supernatural agency exist?
Is the future already decided?
What is the meaning of life?
What is right and wrong?
Is the world good or bad?
Are humans good or evil?
What beings should have what rights?
What should one do?
What is truth?
Consciousness?
Intelligence?
What are the limits of intelligence?
What are the limits of logic?
Could a machine think?
Does free will exist?
How and when did the universe begin?
What happened before it began?
How and when will the universe end?
What does the universe consist of?
What laws govern it?
Why is the universe this way?
How big is the universe?
Does it have a center or edge?
What is outside the universe?
Are there other universes?
What is life?
How did life arise?
What explains its complexity?
How did mind and language arise?
How does the brain work?
Is there life and intelligence beyond earth?
What political system works best?
What economic system works best?
Why do human individuals, groups, and sexes behave as they do?
Why have some human societies experienced more material progress than others?
Will humanity suffer cultural decline?
Economic crash?
Tyranny?
Resource depletion?
Overpopulation?
Runaway pollution?
Pandemic?
Interplanetary impact?
Nuclear catastrophe?
Nanotech plague?
Will humanity experience divine salvation?
Loss of faith?
Paranormal abilities?
Alien contact?
Time travel?
Warp travel?
Machine or human super intelligence?
Immortality?
What will happen in the next: hundred years?
Thousand years?
Million, billion, and trillion years?
The expected results of artificial intelligence
Space travel can be possible as ever before.
Evolution of super-intelligence can be possible.
The task which seems difficult may be completed within no time.
Due to the super-intelligence new complicated discoveries may be possible.
In an artificial intelligence aging and diseases can be eliminated.
Reanimation may be possible (cryonics) in the case of artificial intelligence.
Artificial intelligence minds can be copied easily.
High and efficient labor at low cost.
Basic needs of human beings (food, shelter, cloths) can be fulfilled easily (especially in poor countries).
Communication among the artificial intelligence may be possible (within on time) so that it can help in natural disaster situation.
Artificial intelligence may work completely for the betterment of human race.
They will work as an agent for humans in alien planets (hence it will act as a messenger between humans and aliens or planetary conquest may be possible from alien intelligence).
Crimes may be minimized.
Peace among the nations may be enforced.
The Unexpected results of artificial intelligence
Suppress of human race.
Human race may be treated as slaves to work for the betterment of artificial intelligence.
Human race may also be killed entirely (thinking that we may be a threat to artificial intelligence).
Artificial intelligence may be a competitor for human.
Energy crises may arise exponentially since they need abundant energy resources.
Environmental hazards.
It may be inferior to human race hence it cannot think or work up to the mark of a human (i.e. species whose intelligence is inferior to human like rat, bat etc.).
New crimes may evolve.
It may be use to wage war against other nations.
New types of terrorism may evolve.
Hurdles of artificial intelligence
Till now there are no efficient machines like human which can convert renewable sources like food into pure energy which can be utilized to do work.
Growth cannot be implemented till now (unlike living organisms).
Reproduction may not be possible till now (unlike living organisms).
Repairing of damage parts by itself may not be possible till now (unlike living organisms).
Living organisms find there energy resources or aggregate it by itself (unlike artificial intelligence) till now.
