| 
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • Social distancing? Try a better way to work remotely on your online files. Dokkio, a new product from PBworks, can help your team find, organize, and collaborate on your Drive, Gmail, Dropbox, Box, and Slack files. Sign up for free.

View
 

The Black Swan 

Page history last edited by swanson@... 10 years ago

(Taleb has posted his black swan glossary here)

 

The Triplet of Opacity. Three ailments with history: 

 

 

  1. The illusion of understanding, one thinks he knows what is going on is a random and complicated world.
  2. The Retrospective Distortion, we assess matters after the fact when they are very clear.
  3. The Over Valuation of Factual Information and the handicap of authority when they create categories.  

 

Nassim Nicholas Taleb, the Black Swan

 

He has an example earlier in the book that asks us with some leader forced us to put locks on airplane cabin doors on Sept 10th, 2001 that person would save 1000s of lives but not be celebrated at all. This would be a case in history where we are unable to measure the actual outcome of the action. 

 

Induction Problem: the many problems of making wider theory (predicting future events) out of individual instances

 

2 examples: 1) think of 1000 people (or more). we weigh them. the largest person there will only be a super small fraction of the total weight. Event the most obese person would only be a small fraction. (no black swans) 2) take 1000 people add Bill Gates and compare incomes. Gates' income will make up 99% of the total income. (black swan) When we make generalizations, we need to understand the scope. What are the extremes? What are the outliers that cannot be predicted? 

Another example: Turkey lives 1000 days and is happy and well fed. Day before thanksgiving, he is killed. 

 

Problems of "learning from the past, present, or scene." We assume we can live by things that we observe as being likely. 

Causes of Black Swan (induction problem)

1) error of confirmation: Focus on pre-selected data, what we see and ignore what we do not see

2) Narrative Fallacy:  Fool ourselves by telling stories of simple pattern (we constantly interpret and attribute cause)

3) Human nature looks for patterns and consistency: Act like Black Swan does not exist

4) Distortion of Silent Evident: History hides black swans, shows us things that are not there or show does not show us things that cause events to unfold.

5) Tunnel: We focus on a few, well-defined sources of uncertainty, ignoring other black swans.  

 

(He cites Bertrand Russel and Hume, both of whom drew from predecessors, as key thinkers on this subject. See Sextus Empericus.) 

 

Idea of Falsification: (Problem of falsification) from Popper

He cites a study where people are given the following numbers: 4, 6, 8. They are asked to discover the rule behind this sequence. They have ask about the next numbers in the series. People typically ask if they are 10 12 14, which are correct.  They think the rule is that they are even numbers. But the real rule is that they are increasing. People don't pick enough number sets to see where they get a false response. 

Falsification is the only point that we can get an accurate representation of reality. A series of positive observations doesn't give us a complete view until we see a false case. We don't really know what is going on until we have falsification. So, a single positive statement (I see a white swan), cannot prove a universal (therefore, all swans are white). But, a single observation (I see a black swan), can disprove a universal (all swans are white). 

 

see: Modus Tollens

see: Raven Paradox

 

Silent Evidence

The idea that in many cases the evidence that exists does not represent the overall population of participants since only successful participants produce evidence. Those that fail leave no evidence. He uses the following examples:

 

  • People who pray to survive during emergencies
  • Evolution
  • Ideas like beginners luck
  • most of history 

One thing that I like would be someone who just got lucky gambling who then walks out and claims that gambling is good for the human species because he made all kinds of money.

 

Innovation

"It is hard to think about the car or the computer as being the result of an aimless process, but they are." (need to verify this quote)

He talks about lasers, detecting background radiation in the universe, and other examples of accidental discovery, not planned, purposeful creation. He talk about all of the ways that we use the computer that were not in the realm of thought at all when the computer was first created.

"Knowledge does not progress from tools designed to verify or help theories, but rather the opposite" (p. 169). Theories come from accidents and randomness.

Taleb notes Louis Pasteur's famous quote, "Luck favors the prepared." Taleb notes that Pasteur was great at collecting opportunities. 

"To predict the spread of technology implies predicting a large element or fads and social contagion, which lie outside of the objective utility of the technology itself (assuming there is such an animal as objectivity utility)" (p. 170).

"Prediction requires knowing about technologies that will be discovered in the future. But that very knowledge would almost automatically allow us to start developing those technologies right away. Ergo, we do not know what we will know" (p. 173). 

Law of iterated expectations: "if I expect to expect something at some date in the future, then I already expect that something at present" (p. 172). --strong form

"But there is a weaker form of this law of iterated knowledge. It can be phrased as follows: to undertand the future to the point of being able to predict it, you need to incorporate elements from this future itself. If you know about the discovery you are about to make in the future, then you have almost made it" p. 172). 

 

Experts

Some areas of expertise take "technical" knowledge and some are just knowing stuff--> Greek epistemi vs techni, it is the difference between how and what. 

The question about prediction by experts depends on the type of knowledge and on the range of variation of the subject-->how likely are black swans, how closed is the system, how does the system follow "the model"-->many systems are just too random. there are many times when experts cannot predict better than chance or better than the average person. 

"For many people knowledge has the remarkable power of producing confidence instead of measurable aptitude." 

Experts can be good at predicting regular things, but not always at irregular things. 

"law of iterated expectations"

 

 

Knowledge and Randomness 

In practice, the difference between incomplete information (he calls opacity) and randomness is inconsequential. He talks about using "deterministic chaos" (chaos theory) to understand the world.   There are some realities that are too complex to possibly know until they have already happened...if then. 

The trick is to know history without theorizing about it. Don't make big scientific claims. 

 

 

Kurtosis Risk (fat tail): when the distribution clusters at extremes and does not follow the normal distribution.

Hindsight Bias: 

Availability heuristic: quote from Wikipedia: "Essentially the availability heuristic operates on the notion that "if you can think of it, it must be important."[1] Media coverage can help fuel a person's example bias with widespread and extensive coverage of unusual events, such as airline accidents, and less coverage of more routine, less sensational events, such as car accidents."

I may want to look into the work on prediction of  psychologists Amos Tversky and Daniel Kahneman. 

Postdiction: Predicting events after they have happened. (a critical stance to super natural beliefs)

 

Hedgehog vs Fox: Hedgehog knows one thing, and the fox knows many. Be a fox. (From Tetlock: I may want to look at Philip E. Tetlock's works on experts)

 

In predicting "variability matters"--what is the degree of error of the prediction? Further we do int the future the higher this degree of error. Predictions in many cases (like construction) are only the best case scenarios and do not account for random obstacles that alter their validity. 

Comments (0)

You don't have permission to comment on this page.