DeepMind – YouTube has all the games here. Yes, a computer program run by a powerful computer has defeated humans in the last bastion of human-superiority at playing logic games! I have played Go in the past and it is a game that men excel in and women can, with effort, struggle along but it is obviously harder to do (ahem, I know this for certain) than top male players who tend to be ‘geek’ types. Whenever I did commentary about Go games at the KGS Go Server I used organic comparisons and words to describe ongoing games which proved popular with humans. But computers programmed to play logically tend to make for ‘uninteresting’ boards in my opinion.
How to explain this? Humans see patterns and are very attracted to patterns and impose patterns on everything. A quick look at the planet earth where humans build things called ‘cities’ shows clearly how we need to see and feel patterns and impose this on nature. Self-created patterns are logic systems for moving people and establishing territory that is organic in our minds.
The computer, on the other hand, weighs not strategy or form or shape but ‘future profits’ when leveraging a game choice in the ‘mind’. Humans programmed computers to operate this way and as a person who used to work at a computer circuit system design business many years ago, the final boards created by this human/computer competition reminded me greatly of circuit board designs. More about that later, first, more information for anyone who hasn’t played Go or put together computer boards 50 years ago:
AlphaGo’s win over Go champion Lee Se-Dol proves AI can be ‘immoral’ | Daily Mail Online: HAHAHA. All hail our Deep Mind Overlords! Actually, the danger of computers is, if we program them to ‘fix humanity for x number of reasons’ it will annihilate us all. For example, if you tell Deep Mind Overlords that humans consuming energy are going to make the planet roast us to death, the fix for that would be…kill all humans.
Self-driving cars are already less accident-prone than human drivers, the TV quiz show Jeopardy! is a lost cause, and in chess humans have fallen so woefully behind computers that a recent international tournament was won by a mobile phone.
The first problem is that AI is often trained using a combination of logic and heuristics, and reinforcement learning.
Yes, logic and learning! Many, many humans resist learning anything. We have an infinite capacity to learn nothing from the past. Indeed, we get royally pissed off if we are told, we must stop doing something based on history telling us it is dangerous or futile or stupid. Cassandra’s problem illustrates this: humans will do whatever we want despite all sane warnings. And so it is today: we learn nothing and have very strong will to not learn anything.
Even so, humans have proliferated and taken over most systems which…we then destroy or blow up or let it fall apart or drive it off the cliff, etc.
The logic and heuristics part has reasonably predictable results: we program the rules of the game or problem into the computer, as well as some human-expert guidelines, and then use the computer’s number-crunching power to think further ahead than humans can.
This is how the early chess programs worked. While they played ugly chess, it was sufficient to win.
Reinforcement learning, on the other hand, is more opaque.
Yes, teaching a computer system to be logical is very easy, actually. Computers are, by nature, logic systems. It is all 0 or 1 over and over again, + or -, yes or no choice systems. ‘Thinking ahead’ when playing a ‘logic system’ game is easy for a computer if you give it enough capacity to weigh all possible choice outcomes. This is what I would call ‘how God thinks’.
I grew up around astronomers including my grandparents. I lived literally in observatories for significant parts of my own childhood. My parents and grandparents were very logical about astronomy and utterly illogical about nearly everything else which amused me as a child and also annoyed me considerably. From my child’s mind point of view, my parents would alternately discuss cosmic matters quite clearly and then get all emotional and beat me up if I was too annoying (hint: I was sometimes very annoying!).
When my parents or grandparents were being totally illogical, appealing to logic by explaining it to them was an emotional failure which could increase the rate of beating which is why I got, via the help of Judge Castro, the right to be a full adult when I was only 16 and a half years old. Humans can’t be logical all the time due to the fact that much of our brain is hard wired to be emotional and this shuts down logic systems utterly and totally when hormones (sex, anyone?) and adrenaline are pouring though the brain and the reptile parts where the logic systems are strongest, are shut down so the entity can fight or have sex or go to sleep.
Back to the computer beating the top Go player on earth: the human struggled over and over again, to produce patterns that please on his board while the computer would coldly dissect this and turn it into shapes pleasing to a computer, that is, straight or diagonal structures and not ‘flower-like blossoms’ across the board which I often have seen when watching pros play Go. I watched the online Deep Mind computer playing against the human and it was amusing to watch the human despair as the computer insisted on channeling the game into ‘ugly’ shapes exactly like how computer chess is played by these machines.
From the home base of the computer nerds who built this machine: AlphaGo | Google DeepMind
Go is a popular game in South Korea, China and Japan, and this match was watched or analyzed by hundreds of millions of people worldwide. Many top Go players characterized AlphaGo’s unorthodox plays as seemingly-questionable moves that initially befuddled onlookers, but made sense in hindsight: “All but the very best Go players craft their style by imitating top players. AlphaGo seems to have totally original moves it creates itself.“ AlphaGo appeared to have unexpectedly become much stronger, even when compared with its October 2015 match against Fan Hui where a computer had beat a Go professional for the first time ever without the advantage of a handicap.
I see others have noticed what was clear to me as I watched the games unfold. These moves can be analyzed and I quickly saw that the computer liked to CONNECT FIVE STONES which is what the game is called! Over and over again, whenever the human faltered while building complex shapes, the computer would amuse itself by connecting simply five stones in mostly straight lines, but about 25% of the time, as diagonals.
This is the core program, after all: connect five stones. Period. The human player can force the computer to do other things but the computer more often, forces the human to stick to the Prime Directive: connect those damn stones in groups of five!
…reached an average 80,000 viewers with a peak of 100,000 viewers near the end of game 1.
Viewership: online I know the games attracted more people, maybe over a million. I don’t know. Globally, this is what? 0.01% of the population or less? HAHAHA. Most people don’t know anything about this game but in Asia it is very well known. So humans in the US who didn’t grow up, as I did, around Go games, there is little interest in this news but many computer nerd types who create our computer overlords, Go is a huge thing to them all for obvious reasons.
AlphaGo showed anomalies and moves from a broader perspective which professional Go players described as looking like mistakes at the first sight but an intentional strategy in hindsight. As one of the creators of the system explained, AlphaGo does not attempt to maximize its points or its margin of victory, but tries to maximize its probability of winning. If AlphaGo must choose between a scenario where it will win by 20 points with 80 percent probability and another where it will win by 1 and a half points with 99 percent probability, it will choose the latter, even if it must give up points to achieve it. In particular, move 167 by AlphaGo seemed to give Lee a fighting chance and was declared to look like an obvious mistake by commentators. An Younggil stated “So when AlphaGo plays a slack looking move, we may regard it as a mistake, but perhaps it should more accurately be viewed as a declaration of victory?”
Ahem. DUH. The computer systems looks for best logical outcome not ‘victory’. It doesn’t think, it simply looks at alternate future choices and probability factors before choosing its next move. No rush of emotion, just simple logic systems choices. What many humans forget is, computers do exactly what we tell them to do it is just our way of telling them may confuse us but never the computers. They obey mindlessly.
The programmers simply tapped into larger and larger data processing systems. When I was a very young child, at the University of Chicago, I had a computer learning moment when we were staying in a room (in the process of moving) with my cat, ‘Door kitty’ which I named when this stray cat came meowing to the back door, we were in my father’s office with this kitty cat and he had a key board terminal to a computer processor for entering data.
It was not a modern keyboard, it had 100 buttons on it you had to use to enter the code and I threw the cat on it and it went crazy on me and the cat and I hid under the desk. ‘Do not touch this ever again,’ my father warned me. Traumatic! One of these eventually graduated into being a foot massager under my father’s desk years later!
The thing here is, people who take computer society for granted little think about how new this is, when I was a child, computers were rare and used for things like the space program or planning WWIII. Now, they are in everything, everywhere, all the time. But this entire system is still built on the same logic systems and electronic design protocols that creates things that are logical and systematic and consistent since these things can’t do something that isn’t programmed to happen which is why WWIII will be sudden and utterly successful at annihilating humans since this is the directive given to the computers what will launch this war. Given by humans who programmed this to happen.
Lee apologized for his losses, stating after game three that “I misjudged the capabilities of AlphaGo and felt powerless.” He emphasized that the defeat was “Lee Se-dol’s defeat” and “not a defeat of mankind”. Lee said his eventual loss to a machine was “inevitable” but stated that “robots will never understand the beauty of the game the same way that we humans do.” Lee called his game four victory a “priceless win that I (would) not exchange for anything.”
The poor human, each time he lost, was freaked out. On one occasion, he began to remove the stones and then tried to replace them in a more pleasing human structure and then gave up in despair. He knows in his soul that the final picture of the game was ‘ugly’ and not ‘beautiful’ since it was alien, not human. And this also scared him since the psychology behind the computer’s game was utterly absent.
That is, there really was no ‘beauty’ in any of these games, it was like playing Go with ants or bees. All of which are capable of building very interesting and successful structures and surviving many millions of years of evolution, perfecting this. Unlike my bee colonies, my birds ‘think’ quite differently. All birds I have observed building nests go through the same aesthetic process: they are very, very choosy and look for prime materials of various sorts but will use anything and everything in the end.
But given a choice between, say, some fine lamb’s wool or dirty strips of paper, they will choose the lamb’s wool first. They examine their creations carefully and fuss over how intertwined everything is, etc. But then, in the end, nearly all the nests end up looking nearly exactly the same minus some details. This is the wonder of animal minds at work: we do make many choices based on our ideas of beauty but this is still very much constrained by our own inner Prime Directives.
It’s almost 18 years since IBM’s Deep Blue famously beat Garry Kasparov at chess, becoming the first computer to defeat a human world champion. Since then, as you can probably imagine, computers have firmly cemented their lead over puny, fallible meatbags — Garry Kasparov is still considered by many to be the greatest chess player ever, while computers are only getting more and more powerful. Today, following the completion of TCEC Season 7, we have a new computer chess world champion. Called Komodo, the software can reach an Elo rating as high as 3304 — about 450 points higher than Kasparov, or indeed any human brain currently playing chess.
And the computers relentlessly move forwards as systems for processing data continues to climb, humans are left in the dust. We can’t help it. We cannot out-logic systems that have only logic running for them. Our emotions overtakes our minds. I watched the despair of the human playing against the Go computer and the more the games progressed, the greater his emotional distress whereas the computer was simply relentlessly connecting the stones using simple rules connected with complex choices by foreseeing future outcomes.
The human was emotionally destroyed but the computer would have simply added more data if it lost. No emotions involved at all. And this is what scares humans so much even though we create our own destruction periodically thanks to our wild emotions. We see this clearly today. For example, demands are being made to force European people to take in millions of hostile foreigners who intend to destroy the culture of European people at the same time, the US and EU leaders assist in Jewish genocide against Palestinians and threaten to terrorize or jail anyone who dares talk about this crime against humanity because it is ‘anti-semitism’.
While the EU/US ally, Saudi Arabia, openly supports terrorists who attack Jews and Christians! And we are told over and over again, how good the Saudis are even though they do this. Illogical, of course. Insane, too! And we have an election which has many insane features no computer would tolerate. So, what is our human political prime directive? Is it mindless ‘exterminate’ instructions of the Daleks? Or mindless computer games? Or all of these? We created this, ourselves and at the same time, fear and hate this even though this is our reptilian parts of the brain courtesy of our dinosaur forefathers?
209 Greenhollow Rd
Petersburgh, NY 12138
Make checks out to ‘Elaine Supkis’
Click on the Pegasus icon on the right sidebar to donate via Paypal.