Antifragile is another name for adaptive response, and it is one reason why models fail to be reliable very far into the future. The other reason is that many processes have probabilistic distribution that are not finite which Talib describes as fat tailed. Fat tailed processes have distributions which have no mean, no variance, or higher moments. Yes, you can compute these values, but they are not meaningful, and the resulting models are GIGO.
Failure to recognize adaptive response and fat tailed processes are major reasons why so much stupidity is going on in our world.
Chet
February 12, 2022
Comment to Quote of the day—Forrest Cooper
[This quote probably isn’t fully understood by more than 10%, at best, of the general population. But it really resonates with me and I wanted to give it more visibility so others who understand it can bask in the awesomeness.—Joe]
I am the 90%. Any way to explain it?
Many things, perhaps most, have a ““normal distribution” (bell-shaped) curve. For example, most people are around average height, there are some very tall or very short people, and a tiny number of giants and dwarfs. A lot of statistical measures are very good at describing and predicting these sorts of populations given a only a partial sample of the data. But if something can rapidly change in response to changing pressures in the environment then predicting future distribution curves based on historic ones will lead you very far astray. This is not the same as simply a long tail in a wide, flat distribution curve. A related problem is when you have a population with a huge number of “normals” and a few extreme outliers that skew all your normal stats measures into meaninglessness.
Possible good starting point for it is here (five and a half minutes): https://nassimtaleb.org/2021/04/fat-tails-introduction/
Predictions are almost always of the form all things being equal and continuing, such and such will happen. It is the assumption that all things being equal and continuing that is almost never true. Put simply systems adapt. It’s true in physics and it is certainly true with humans and life in general. And it is stupid for officials to think otherwise. Look at how well the COVID models worked along with the resulting lock downs and mask mandates.
Underlying the adaptive response is the very nature of systems that leads to chaotic behavior. Mathematically almost all our tools and models of these systems are based on the normal distribution because it is tractable and finite. In other words, we can describe the height of humans using a normal probabilistic distribution with a mean and a variance and be accurate over some time scale. However, even with height there are lots of variables that will change, and we know that height has changed even in modern times, and we know that height varies between groups. Still the normal distribution is useful in understanding height of humans. However, many systems change on a much more frequent basis and involve more chaotic behavior.
The weather is one such system. Today we can predict with some accuracy a week to two weeks out. Beyond that weather becomes unpredictable because small changes can result in big changes. A common saying is that a butterfly flapping its wings in China can result in a thunderstorm in Kansas two weeks later. That’s chaos and it is not predictable and certainly not provable for any one case. Mathematical these processes are described as fat tailed distributions that have no theorical central descriptions like a mean or variance. In other words, they can have any value. Look up the stable Cauchy distribution in Wikipedia. Also see http://www.sr-sv.com/the-dangerous-disregard-of-fat-tails-in-quantitative-finance/. And read The Black Swan or Antifragile by Talib.
There are limits to knowledge due to both adaptive response and chaos. And stupidly makes it even worse. Just because tech has been so successful during our generation does not mean that we will likewise be able to solve other problems with equal success. Yet, that is exactly the view of the elite progressives that are destroying the civilization that we have because reality is not acceptable. That goes for sex and gender, and it goes for inequality.
I also forgot to mention the stupid misuse of the tools like data mining that are used to generate headline after headline using the weasel words may, possible, or likely.
Also consider in your calculations of modern behavior based in millennia of “Psycho-history” a guy who still, after about a millennium of women claiming their humanness, refers to evil as “she” instead of “it”.
A very interesting thesis.
I would argue that it is incomplete. “Failure to recognize adaptive response” explains why the results of adaptive response are hard to predict. But a chaotic world needs more explanation than that. I suggest that, among adaptive responders, some of them will be bad adaptive responders; others will be self-serving, and destructive to the whole.
I agree, though, that some systems will adapt quickly — get inside the OODA loop of others, if you will — and will thus be more successful (for however they define success). And others, trapped in a slower loop, will be helpless to explain what the faster systems are doing and why, and will tend to flail in response.
At a certain point, it becomes important, not just to identify who your friends and your enemies are, and in what contexts. (Your enemy in one context may be your friend in another.) It also becomes critical to identify the slow-movers and the fast-movers… for if your friend has a slow OODA loop and your enemy has a fast OODA loop, your friend will not be able to help you much.
We live in sobering times.
“Antifragile is another name for adaptive response.” Which to me is the true measure of intelligence? Someone’s real IQ?
I believe it was Mike Tyson that said, “everyone has a plan till they get punched in the face.”
Which I’m sure was a rip-off of the military axiom that. “No plan survives first contact.”
I would posit that human nature is what drives most systems to go Gila Monster.
As humans need plans to give inspiration to action first and foremost.
And stubborn adherence to it in the face of chaos will prove it’s downfall.
(I would also posit that it is at those points in chaos that some humans are at their best.)
But the main point is that the more man-hours your model has invested. The less likely it will adapt to chaos. (Tower of Babel.)
If you want to hear god laugh, tell him your plan.
Humans need chaos and fat-tails to be better humans. Which is what god wants.
Computer systems are just human tools. Another failure point in human growth.
America is living and dying through one of those human fat-tails right now.
We live in a god created catch-22 if you will. As, no one here gets out alive.
Meaning, the wisdom gained is mostly lost.
The beating will continue until moral improves, will always be life on this planet.
Being anti-fragile is an exercise in mental pain management.
Adaptive response, the measure of intellect.
And this circular logic I’m spewing is invoking chaos on my meager mental processes.
Yes, well don’t discount the notion that the enemy has been studying and experimenting with what we might call “mass human behavior” for a very long time.
Isaac Asimov posited openly back in the 1940s and ‘50s, in his Foundation series, the ideas of “psychohistory” and “mathematical sociology”, but the basic concepts, by other names, go back into antiquity. I say “openly” because if it’s been said openly then it’s been around for some time before, and furthermore we have to assume there’s been a lot of less open investigation into the concept since.
What we may think of as “antifragility” and “fat-tailedness” aren’t necessarily as incalculable as we prefer to think. They’re certainly not new concepts. Consider the world as a test ground for various theories, an experimental laboratory for fleshing out various mass responses to various imposed scenarios (the 20th Century for example should be understood as one big experiment). The continuous conflict in Middle East in the last 30 years has no doubt left the enemy with a lot of good data along the very lines of inquiry that we’re discussing here.
And so the kinds of veiled and semi-veiled allusions to guerrilla warfare, or adaptive and spontaneous yet organized and unconventional counter-insurgency, avoidance, evasions, work-arounds, et al, are, I submit, already part of the formulae.
I keep saying it— Don’t make the mistake of believing that the enemy is dumb, and hasn’t taken into consideration the many and various forms of push-back, adaptive response, etc., that we may have in mind as being “unpredictable”. I now consider the enemy to be quite brilliant. Evil of course, but brilliant, even in her copious and public use of the blind and stupid.
And don’t think that the enemy hasn’t dealt with our kind before. She has. All over the world, across many cultures and generations. It may be therefore, that in our prideful, eager delusions of unpredictability we are, on some level, quite thoroughly predictable. Consider these things in your calculations.
Between Hall and Bernays. Marx and Engels. Methinks this game has been being played since 1850’s. Right along with communism. The theory of evolution by natural coincidence. And Keynesian economics.
It’s almost like someone was paying them to come up with this crap?
Oh, it’s much older than that. The central power of the Dark Ages is still with us, for example, and still wields power, and it’s been gathering knowledge and experience, and perfecting its power base and its network of associations, for ~1.5 millennia, and its roots go back much further still.
Someone like the PM of Canada today for a current and local example doesn’t matter a whit. He’s a puny little whore, and if the citizens get rid of him there’ll be another puny whore already prepped and ready to take his place. Same goes for other things which we may think of as critical assets of the enemy. They’re like chess pieces which may be easily sacrificed to lure an opponent into a checkmate, except that our enemy has a virtually inexhaustible supply of those chess pieces!
What we generally fail to understand, and often refuse to believe even when the evidence is staring us in the face and shouting, is that the enemy is global, it thinks and understands much deeper than we do, it considers things we would consider unthinkable, and plays the long game. The details of what happens in this or that country or region therefore are of little significance so long as the attentions and mindsets of the people are sufficiently affected and directed.
And of course I know very well how crazy that all sounds. Too bad. It’s the very reason, by the way, that we cannot perceive it. Like the German citizens of the 1930s and ‘40s, we ourselves simply cannot wrap our heads around the kind of evil that’s gotten its beachheads established and is poised over the entire world today.
Pingback: Quote of the day—Igor Volsky | The View From North Central Idaho