Modelling Seed A.I. on the Brain
Seed AI is one of the hopes and dreams upon which AI is based, the question asked within this article (and hopefully, answered) is; What do we need for it ?
AI as it now stands is capable of navigation, light conversation, vision and pretty much anything else ending in 'ion'. However, complain about a squeaky car door to a medical system, like Generation5's MARS, and expect to find yourself being told to buy some cod-liver oil tablets. It isn't a flexible system, processing information can only be done through certain channels. This is the goal of Seed AI, to be flexible, before we see how, we must understand how.
Take a newborn baby, and put him/her/it (ignore as appropriate) into a pitch-black room. Within a day, that baby will lose all neurological capacity for vision. When you take them out, they'll see again, after a short while. This is because the brain is the most flexible system ever developed, Seed AI will have to be just as flexible.
Take a blind man, about 30 or so years old, who has been blind throughout his life, and, through faith, or just good surgery, allow him to see for the first time. What do you think will happen? That he can see? That he can move about unaided? WRONG!
He'll be unable to understand distance, unable to understand colour. He's never experienced it before, and his brain isn't as flexible as he was when he was a baby. Intellectually he might be able to understand that some objects have an intervening space between them. He just wont understand how! His mind isn't as flexible, no, but it is more refined to what it has experienced.
Some people have laughed at the idea of Seed AI, in my own view, this is much like putting a book of Quantum Mechanics in front of a 3 year old, and expecting them to come up with an essay about it, preferably without badly drawn characters in it. It's nigh impossible to happen, and takes patience. You yourself's, I imagine, didn't come out the womb asking to be taken real estate shopping.
Coming away from the bed-time stories, we go straight into neurological facts:
- 6 Billion neurons in your head.
- Each individual neuron is connected up to 100,000 others.
- Each individual neuron has about 2400 Micro-tubules inside it.
- Putting one neuron in front of the other, you'd end up with a piece of 'string' 50KM long!
I don't think you have to specialize in computational neuroscience to realize that something as complicated as the brain, remade with today's technology, is going to be about the size of a house.
Glial cells, as far as I've seen, have been overlooked when it comes to making artificial brains. Gilal cells are the 'workforce' of the brain, they protect it and serve it. Think of it as the handy-man in the Palace of the Mind. They support, provide nutrition (as the blood, unlike all the other organs in the body, never comes into contact with neurons itself), maintain homoeostasis, remove viruses and other toxic material, and get rid of dead neurons. Granted, if you've ever heard of dead brain tissue being a problem, its because its often then not too large for the glial cells to get rid ofin the amount of time before it swells up and causes damage. They hold the neurons together and help with synaptic transmission, clearing away any excess chemicals. When Neurogenesis starts, they guide the new neurons into position. They outnumber neurons 10 to 1. Yet no-one seems to recognize the importance of them!
EURISKO was probably the first step with A.I, simply because the below article can get the concept across better then I, I decided it would be better to cut and paste (with a copyright included, of course) and leave you to understand the whole mess better ;).
On the July 4 weekend of 1981, while many Americans were preoccupied with barbecues or fireworks displays, players of an immensely complex, futuristic war game called Traveller gathered in San Mateo, California, to pick a national champion. Guided by hundreds of pages of design rules and equipment specifications, players calculate how to build a fleet of ships that will defeat all enemies without exceeding an imaginary defense budget of one trillion credits.
To design just one vessel, some fifty factors must be taken into account: how thick to make the armor, how much fuel to carry, what type of weapons, engines, and computer guidance system to use. Each decision is a tradeoff: a powerful engine will make a ship faster, but it might require carrying more fuel; increased armor provides protection but adds weight and reduces maneuverability.
Since a fleet may have as many as 100 ships–exactly how many is one more question to decide–the number of ways that variables can be juxtaposed is overwhelming, even for a digital computer. Mechanically generating and testing every possible fleet configuration might, of course, eventually produce a winner, but most of the computer’s time would be spent blindly considering designs that are nonsense. Exploring Traveller’s vast "search space," as mathematicians call it, require the ability to learn from experience, developing heuristics–rules of thumb–about which paths are most likely to yield reasonable solutions.
In 1981, Eurisko, a computer program that arguably displays the rudiments of such skills, easily won the Traveller tournament, becoming the top-ranked player in the United States and an honorary Admiral in the Traveller navy. Eurisko had designed its fleet according to principles it discovered itself–with some help from its inventor, Douglas B. Lenat, an assistant professor in Stanford University’s artificial-intelligence program.
"I never did actually play Traveller by hand," Lenat said, three years later. "I don’t think I even watched anybody play it. I simply talked to people about it and then had the program go off and design a fleet…When I went into the tournament that was the first time that I had ever played the game." Eurisko’s fleet was so obviously superior to those of its human opponents that most of them surrendered after the first few minutes of battle; one resigned without firing a shot.
Eurisko makes its discoveries by starting with a set of elementary concepts, given to it by a human programmer. Then, through a process not unlike genetic evolution, it modifies and combines them into more complex ideas. As structures develop, the most useful and interesting ones-judged according to standards encoded in the program-survive.
At the time of the Traveller tournament, Lenat had already used a forerunner of Eurisko to grow mathematical concepts, getting the program to rediscover arithmetic and some theorems in elementary number theory. Now the structures Lenat wanted to see evolve were Traveller fleets. He provided the program with descriptions of 146 Traveller concepts, some of them as basic as Acceleration, Agility, Weapon, Damage, and even Game Playing and Game. Others were more specific: Beam Laser, Meson Gun, Meson Screen, and Computer Radiation Damage.
A Eurisko concept can be thought of as a box containing "slots" filled with information describing it. For example, the "Is A" slot in the box representing Energy Gun indicates that it is a Defensive Weapon Type and an Offensive Weapon Type–and a Physical Game Object, as well. These concepts are, in turn, described by other boxes. Another slot tells Eurisko that information on Energy Gun’s firing range will be found in a box called Energy Gun Attack Info.
With a network of these boxes interlinked in its memory, Eurisko began designing ships and simulating battles. After each altercation, it analyzed the results, made adjustments to the fleets and tried the battle again. In the process, Eurisko tested Traveller concepts by natural selection. For example, after a number of battles, Eurisko discovered how easy it was to provide ships with enough armor to protect them against energy guns. Thus the value in the Worth slot of Energy Gun, which was originally set at 500, was eventually lowered to 100. Weapons that proved more valuable would increase in worth, toward a maximum value of 1000. Gradually an ever-more-invincible Traveller fleet evolved.
"At first," Lenat later wrote, "mutations were random. Soon, patterns were perceived: more ships were better; more armor was better; smaller ships were better; etc. Gradually, as each fleet beat the previous one…its lessons were abstracted into new specific heuristics."
When Eurisko began its experiments, the My Creator slot in each of its concepts all contained the name Lenat. But, as Eurisko played, an increasing number of the slots were filled with the name of the heuristic that had been used to synthesize them.
Eurisko was creating concepts on its own. It was distilling thousands of experiences into the judgmental, almost intuitional, knowledge that constitutes expertise–rules that can’t always be proved logically, that can’t guarantee a correct answer, but that are reliable guides to the way the world works, a means of cutting through complexity and chaos.
Each night, Lenat would leave Eurisko running on a personal computer in his office, returning in the morning to read the results, occasionally helping the process by encouraging the conjectures that seemed most fruitful and weeding out mistakes.
"Thus the final crediting of the win should be about 60/40% Lenat/Eurisko," he wrote, "though the significant point here is that neither party could have won alone. The program came up with all the innovative designs and design rules…and recognized the significance of most of these. It was a human observer, however, (the author) who appreciated the rest, and who occasionally noticed errors or flaws in the synthesized design rules which would have wasted inordinate amounts of time before being corrected by Eurisko."
After weeks of experimentation, and some 10,000 two-to-thirty-minute battles, Eurisko came up with what would be the winning fleet. To the humans in the tournament, the program’s solution to Traveller must have seemed bizarre. Most of the contestants squandered their trillion-credit budgets on fancy weaponry, designing agile fleets of about twenty lightly armored ships, each armed with one enormous gun and numerous beam weapons.
Eurisko, however, had judged that defense was more important than offense, that many cheap, invulnerable ships would outlast fleets consisting of a few high-priced, sophisticated vessels. There were ninety-six ships in Eurisko’s fleet, most of which were slow and clumsy because of their heavy armor. Rather than arming them with a few big, expensive guns, Eurisko chose to use many small weapons.
In any single exchange of gunfire, Eurisko would lose more ships than it destroyed, but it had plenty to spare. The first battle in the tournament was typical. During four rounds of fire, the opponent sank fifty of Eurisko’s ships, but it lost nineteen–all but one-of its own. With forty-six ships left over, Eurisko won.
Even if an enemy managed to sink all Eurisko’s sitting ducks, the program had a secret weapon–a tiny, unarmed extremely agile vessel that was, Lenat wrote, "literally unhittable by any reasonable enemy ship." The usefulness of such a ship was discovered during a simulated battle in which a lifeboat remained afloat round after round, even though the rest of the ships in the fleet had been destroyed. To counter opponents using the same strategy, Eurisko designed another ship equipped with sophisticated guidance computer and a giant accelerator weapon. Its only purpose was killing enemy lifeboats.
After Eurisko prevailed so easily, the tournament’s directors tried to ensure that the 1982 championship would be different."They changed the rules significantly and didn’t announce the final new set of rules until a week or so before the next tournament," Lenat said. "The first year that would have not been enough time for me to run the program to converge on a winning fleet design." But Eurisko had learned heuristics that were general and powerful enough that they could be applied to new versions of the game.
"We won again and they were very unhappy and they basically asked us not to compete again. They said that if we entered and won in 1983 they would discontinue the tournaments. And I had no desire to see that happen." So Eurisko retired undefeated.
©1984 George Johnson