Twitter: @rodgermitchell; Search #monetarysovereignty
Facebook: Rodger Malcolm Mitchell
In the post, “What is the illusion of intelligence,” we discussed Artificial Intelligence (AI), and said that our rudimentary computers and lower animals (ants) have “intelligence.”
Reader “scottonthespot” demured, saying in part:
There is no such thing as Artificial Intelligence. It is all just Simulated Intelligence.
Some programming to make it LOOK like the machine is acting like us, motivated by the same things we are. But it is no more real than the animatronics at Disney World.
No computer has true drive, desire, hunger, fear, want, joy, anger, or any of the other complex emotions that are built into our non-conscious brain.
In fact, a single-celled amoeba has more in common with us than IBM’s Watson, even though we’ve programmed the latter to SEEM like us.
The amoeba, like us, gets hungry, flees from danger (in its limited way), desires to reproduce (albeit asexually), etc.
It took about half a billion years to get from that to multi-celled animals and then relatively quickly to complex animals like us – with a whole bunch of fairly arbitrary extinctions along the way.
None of this was planned from the top-down in the way we are trying to do with so-called A.I.
Show me a single computer anywhere that actually fears being turned off – and doesn’t just mimic our fears of death in words and actions.
A.I. Researchers have basically given up on their earlier quest to create intelligence like ours.
Who knows? Big Data may prove more useful. But it isn’t intelligence. It’s just a simulation.
My response was:
No law says “intelligence” must resemble the human brain or have involved an evolution that replicates human evolution.
Further, there are specific parts of the human brain having to do with fear, joy, anger, etc., and there actually are living individuals who do not have one or more of these emotions, because of damage to certain brain cells — yet they have intelligence.
Anyway, we’re talking about “artificial” intelligence rather than duplicating human intelligence, and artificial intelligence may become smarter in many ways than human intelligence.
It depends on the criteria with which you measure intelligence. Even my iRobot Roomba cleaner demonstrates hunger, and it flees from danger.
When it’s hungry it goes to the electrical outlet, and if it’s in danger of falling down the stairs, it moves away.
And yes, even my little desktop computer has “fears.” It asks me, “Are you sure?” when I want to leave certain pages. Some computer viruses actually prevent shutdown unless you remove all power, and even then, when you power up again the same page may reappear.
But “scottonthespot” was correct when he wrote, “None of this was planned from the top-down in the way we are trying to do with so-called A.I.” And in fact, the “top-down” method, in which programmers try to create algorithms that duplicate thinking, simply does not work for complex problems like face or voice recognition.
NewScientist Magazine: 26 November 2016: If you compare two passport photos of yourself taken one minute apart, they will not be identical at the level of raw pixels.
This is sufficient for the computer to treat them as two completely different images.
We would like the computer to represent those images in a more robust way than just using pixels, so it does not get confused by small irrelevant changes in the images.
Programming this capability directly into a computer has proven difficult, so engineers have resorted to machine learning.
All of AI has moved on to “machine learning,” which actually parallels human intelligence far better than does the “top down” method.
I never have had a completely original thought — nor have you. Our every thought, no matter how creative, is based on all our life’s experiences. I cannot visualize colors beyond ultraviolet simply because I never have seen colors beyond ultraviolet. No amount of creative effort will change this.
I never will write haiku in Japanese. I don’t know Japanese. No amount of intelligence will change this.
There are many measures and descriptions of intelligence and creativity, but any measure or description requires background data and some form of reassembly.
Everything we think, believe or imagine has devolved from some prior experience, which is exactly how machine learning operates.
You decide on the right thing to wear tonight. Your decision is based on a combination of historical data — all the data from your many, many trillions of life’s second-by-second experiences — subconsciously assembled via repetition and probability.
You cannot explain exactly how you came to your decision. You just “feel” it via an internal algorithm, which you cannot explain. This is “bottom up” intelligence. This is human intelligence.
Machine learning, like you, assembles trillions of historical data points, and via repetition and probability, comes to a mathematical “best-fit” conclusion.
For face recognition, machine learning involves showing the computer trillions of images, and telling the computer which ones depict the same faces. Over time, the computer begins to “get it,” by creating its own internal algorithms.
Even when the pixels may differ substantially, the computer has “learned” which differences are important. No human could write such algorithms, though we use similar algorithms subconsciously every time we recognize a face.
Which brings us to economics:
Every economist produces “if/then” hypotheses. If interest rates go up, then this will happen. If the minimum wage goes up then that will happen. And if deficits go up, then another thing will happen.
And we repeatedly are wrong, especially in the details.
I can tell you, with much assurance, that if the federal government were to run a surplus, or merely to reduce deficit growth (as in a balanced budget), we would have a recession. I know this because it has happened many times before — history, repetition, probability.
But I can’t tell you how soon, how deep, and how long the recession will be. No one can.
Stock market chartists try to predict future market moves by analyzing past market moves, but they too often are wrong, being unable to factor in the underlying causes of past market moves.
Machine learning will do this. “Big data” will compile all past market changes, together with data about world currencies, GDP, censuses, leaderships, politics, technology, energy, discoveries, wars, weather, ecology, and other factors.
There need be no assumptions. It all will be data and probability.
No human could write an algorithm that could take into account so many factors, but the machine does, not by “intelligence” in the way we think of it, but rather by running those numbers again and again until it arrives at a “best fit.”
The machine does not “know why” its answer is correct. It just spits out the the most mathematically probable answer — which would be close to the way human reasoning and intuition operate, if we were able to handle as much data as a machine can.
Given sufficient data and computing power, machines will be able to predict the future far better than we humans can.
Not only will they be better able to predict the future based on today’s reality, but they will be better able to predict “if/then” futures.
Using big data and machine learning, they will answer such questions as:
“If we pass a Balanced Budget Amendment today, then how soon will the depression probably begin; how deep will it probably it go; how long will it probably last?” What are the best fit probabilities?
And, “If the U.S. President is an egomaniacal, mean-spirited liar, who cares nothing for the people, but is interested solely in using the Presidency to grow his own wealth and glory, and if he appoints similarly mean-spirited incompetents as advisors, then what are the best-fit probabilities of what will happen to the “99%” of America?”
That will be the beginning of economics as a real science.
Rodger Malcolm Mitchell
•Those, who do not understand the differences between Monetary Sovereignty and monetary non-sovereignty, do not understand economics.
•Any monetarily NON-sovereign government — be it city, county, state or nation — that runs an ongoing trade deficit, eventually will run out of money.
•The more federal budgets are cut and taxes increased, the weaker an economy becomes..
•No nation can tax itself into prosperity, nor grow without money growth.
•Cutting federal deficits to grow the economy is like applying leeches to cure anemia.
•A growing economy requires a growing supply of money (GDP = Federal Spending + Non-federal Spending + Net Exports)
•Deficit spending grows the supply of money
•The limit to federal deficit spending is an inflation that cannot be cured with interest rate control. The limit to non-federal deficit spending is the ability to borrow.
•Until the 99% understand the need for federal deficits, the upper 1% will rule.
•Progressives think the purpose of government is to protect the poor and powerless from the rich and powerful. Conservatives think the purpose of government is to protect the rich and powerful from the poor and powerless.
•The single most important problem in economics is the Gap between the rich and the rest.
•Austerity is the government’s method for widening the Gap between the rich and the rest.
•Until the 99% understand the need for federal deficits, the upper 1% will rule.