Hide table of contents

This is a series of posts for the Bayesian in a hurry. The Bayesian who wants to put probabilities to quantities, but doesn't have the time or inclination to collect data, write code, or even use a calculator.

In these posts I'll give a template for doing probabilistic reasoning in your head. The goal is to provide heuristics which are memorable and easy to calculate, while being approximately correct according to standard statistics. If approximate correctness isn't possible, then I'll aim for intuitively sensible at least.

An example of when these techniques may come in handy is when you want to quickly generate probabilistic predictions for a forecasting tournament.

In this first post, I'll cover cases where you have a lower bound for some positive value. As we'll see, this covers a lot of real-life situations. In future posts I'll use this as a foundation for more general settings.

For those who struggle with the mathematical details, you may skip to the TL;DRs.

The Delta T Argument

We'll use what John Gott called it the "delta t argument" in his famous Doomsday Argument (DA). It goes like this:

  1. Suppose humanity lasts from time  to time  .
  2. Let  be the present time, and  be the proportion of humanity's history which has so far passed.
  3.  is drawn from a uniform distribution between zero to one. That is, the present is a totally random moment in the history of humanity.
  4. The probability that  is less than some value  is

5. Let . Then humanity's survival function (the probability of humanity suviving past time ) is hyperbolic:

6. And the density function for human extinction is given by

These distributions are illustrated below:

If we choose time units so that Tp=1, then humanity's survival function is given in green, and the extinction  density function is given in red.

TL;DR The probability of a process surviving up to a certain time is equal to the proportion of that time which has already been survived. For example, the probability of a house not catching fire after 40 years, given it has so far lasted 10 years without catching fire, is .

Lindy's Law

Lindy's Law states that a process which has suvived up to time , will on average survive a further . This is true for the delta t distribution, as long as we interpret "average" to mean "median":

However, the mean lifespan of humanity is undefined:

On a related note, because our density function decreases with the inverse square of time, it is "fat tailed", meaning that it dies off sub-exponentially. It belongs in Nassim Taleb's Extremistan.

TL;DR A process which has survived  amount of time, will with 50% probability survive another  amount of time. For example, a house which has lasted 10 years without burning down, will survive another 10 years without burning down with 50% confidence.

50% Confidence Intervals

We can thus derive a 50% confidence interval by finding the  values at which humanity has a 25% chance of surviving, and a 75% chance of surviving. The first is obtained from

which when we solve for  gives

The second is obtained from

which gives

So with 50% confidence we have

which we can remember with the handy mnemonic:

Adding a third is worth a lower quartile bird.

Times-ing by four gets your last quartile in the door.

In the case of human extinction: Homo Sapiens have so far survived some 200,000 years. So with 50% confidence we will survive at least another 60,000 years, and at most another 800,000 years.

TL;DR A process which has survived  amount of time, will with 50% confidence survive at least another  and at most another  amount of time. 

90% Confidence Interval

By similar reasoning, we can say with 90% confidence that

But it's hard to muliply by 20/19. So instead we'll approximate it with

Thus, with about 90% confidence, the remaining lifespan is more than 1/20th of its current lifespan, but less than 20x its current lifespan. 

Similarly, we can say that with about 99% confidence the remaining lifespan is more than 1/200th the current lifespan, but less 200x times the current lifespan. And so on with however many nines you like.

So with 90% confidence, humanity will survive at least another 10,000 years, and at most another 4,000,000 years, and with 99% confidence between 1,000 and 40,000,000 further years.

TL;DR A process which has survived  amount of time, will with 90% confidence survive at least another  and at most another  amount of time. 

The Validity of the Doomsday Argument

A lot of people think the DA is wrong. Should this concern us?

I think in the specific case of predicting humanity's survival: yes, but in general: no.

 When you apply the delta t argument to humanity's survival you run into all kinds of problems to do with observer-selection effects, disagreements about priors, disagreement about posteriors, and disagreement about units. For an entertaining discussion of some of the DA's problems, I recommend Nick Bostrom's The Doomsday Argument, Adam & Eve, UN⁺⁺, and Quantum Joe.

But when you apply delta t argument to an everyday affair, such as the time until your roof starts leaking, then you needn't worry about most (or all) of these problems. There are plenty of situations where "the present moment is a random sample from the lifetime of this process" is a perfectly reasonable characterisation.

Let's look at some everyday examples.

Examples

Example 1: Will a new webcomic be released this year?

Your favourite webcomic hasn't released any new installments since 6 months ago. What is the probability of a new installment this calendar year (ie, within next 9 months)?

Answer: The probability of the no-webcomic streak continuing for a further 9 months is

Which is a little over a third. Maybe 35%. So the probability of there being a comic is about 65%.

Example 2: How high will my balloon chair go?

You have tied 45 helium balloons to a lawnchair and taken flight. You are now at an altitude of 1,000km. How high will you go?

Answer: Our 90% confidence interval gives at least 1,200km and at most 20,000km. Lindy's Law gives an average height of 2,000 km.

Example 3: Will the sun rise tomorrow?

I have observed the sun rise for the last 30 years. What is the probability that the sun rises tomorrow?

Answer: 3 years is about 1,000 days, so 30 years is about 10,000 days. The probability that the sun rising streak ends on the 10,001st day is 

Which itself is going to be approximately 1/10,000=0.001%. So the sun will rise with 99.999% confidence.

"Real" answer: Laplace's answer to the sunrise problem was to start with a uniform prior over possible sunrise rates, so that the posterior comes out as

This is the "rule of succession", and which in our case also gives something very close to 0.001%. Alternatively, we could use the Jeffreys prior and get 

which will be something more like 0.0015%. 

Example 4: German Tanks

You have an infestation of German tanks in your house. You can tell they're German because they're tan with dark, parallel lines running from their heads to the ends of their wings. You know that the tanks have serial numbers  written on them. You inspect the first tank you find and it has serial number . How many tanks are in your house?

Answer: 120-2,000 with 90% confidence. Median is 200.

"Real" answer: If we are doing frequentist statistics, the the minimum-variance unbiased point estimate is , so 199. The frequentist confidence intervals are gotten by the same formula as the delta t argument, so we again have 120-2,000 with 90% confidence.

The Bayesian story is complicated. If we have an improper uniform prior over , then we get an improper posterior. But if we had inspected two tanks, and the larger serial number was 100, then we would have a median estimate of  (the mean is undefined). If our prior of  was a uniform distribution between 1 and and upper bound , then the posterior looks like

which has approximate mean

So if we have an a priori maximum of  tanks, then the mean will be something like . Weird.

I don't know what happens if you use other priors, like exponential.

Next time...

In future posts in this series I'll cover situations where you need estimating a distribution from a single data point.

Comments3


Sorted by Click to highlight new comments since:

I really like the idea behind this post/series. I'd already come across Lindy's Law/delta T and the rule of succession, by reading other people use it in their predictions, but I had already thought that this was a really inefficient way to learn. I skimmed a few statistics textbooks, but I did not come across a lot of techniques that I actually ended up using. 

I also liked the examples you gave. I felt like 1-3 explicit practice Problems at the end would also have been nice like:

Tesla was founded in 2003.

  • How many years from now does tesla have a 25/75% chance to exist?

Or maybe this is silly?

Anyway...

I knew that the Lifetime of something depends on the time it stuck around and had a rough mental image of the distribution, but so far I did not actually bother calculating it explicitly. So thanks for the heuristics.

Your post actually made me think about how very often the lifetime of something is very dependent on the lifetime of something else whose distribution is better known. Often you can just substitute one probability for the other, but sometimes this is more difficult. For example, when someone is 60 and he has been in the same company for 45 years then I don't expect him to stay another 45, because I roughly know when people tend to retire which in turn is dependent on the expected lifetime of someone. The most extreme/ridiculous form of this is of course how every long-term forecast you make can be totally dominated by your timelines for AGI.

Nice post! Found it through the forum digest newsletter. Interestingly I knew Lindy's Law as the "Copernican principle" from Algorithms to Live By, IIRC. Searching for the term yields quite different results however, so I wonder what the connection is.

Also, I believe your webcomic example is missing a "1 -". You seem to have calculcated p(no further webcomic will be released this year) rather than p(there will be another webcomic this year). Increasing the time frame should increase the probability, but given the formula in the example, the probability would in fact decrease over time.

Thank you. I have corrected the mistake.

The relationship between Lindy, Doomsday, and Copernicus is as follows:

  • The "Copernican Principle" is that "we" are not special. This is a generalisation of how the Earth is not special: it's just another planet in the solar system, not the centre of the universe.
  • In John Gott's famous paper on the Doomsday Argument, he appeals to the the Copernican Principle to assert "we are also not special in time", meaning that we should expect ourselves to be in a typical point in the history of humanity.
  • The "most typical" point in history is exactly in the middle. Thus your best guess of the longevity of humanity is twice its current age: Lindy's Law. 
Curated and popular this week
 ·  · 25m read
 · 
Epistemic status: This post — the result of a loosely timeboxed ~2-day sprint[1] — is more like “research notes with rough takes” than “report with solid answers.” You should interpret the things we say as best guesses, and not give them much more weight than that. Summary There’s been some discussion of what “transformative AI may arrive soon” might mean for animal advocates. After a very shallow review, we’ve tentatively concluded that radical changes to the animal welfare (AW) field are not yet warranted. In particular: * Some ideas in this space seem fairly promising, but in the “maybe a researcher should look into this” stage, rather than “shovel-ready” * We’re skeptical of the case for most speculative “TAI<>AW” projects * We think the most common version of this argument underrates how radically weird post-“transformative”-AI worlds would be, and how much this harms our ability to predict the longer-run effects of interventions available to us today. Without specific reasons to believe that an intervention is especially robust,[2] we think it’s best to discount its expected value to ~zero. Here’s a brief overview of our (tentative!) actionable takes on this question[3]: ✅ Some things we recommend❌ Some things we don’t recommend * Dedicating some amount of (ongoing) attention to the possibility of “AW lock ins”[4]  * Pursuing other exploratory research on what transformative AI might mean for animals & how to help (we’re unconvinced by most existing proposals, but many of these ideas have received <1 month of research effort from everyone in the space combined — it would be unsurprising if even just a few months of effort turned up better ideas) * Investing in highly “flexible” capacity for advancing animal interests in AI-transformed worlds * Trying to use AI for near-term animal welfare work, and fundraising from donors who have invested in AI * Heavily discounting “normal” interventions that take 10+ years to help animals * “Rowing” on na
 ·  · 3m read
 · 
About the program Hi! We’re Chana and Aric, from the new 80,000 Hours video program. For over a decade, 80,000 Hours has been talking about the world’s most pressing problems in newsletters, articles and many extremely lengthy podcasts. But today’s world calls for video, so we’ve started a video program[1], and we’re so excited to tell you about it! 80,000 Hours is launching AI in Context, a new YouTube channel hosted by Aric Floyd. Together with associated Instagram and TikTok accounts, the channel will aim to inform, entertain, and energize with a mix of long and shortform videos about the risks of transformative AI, and what people can do about them. [Chana has also been experimenting with making shortform videos, which you can check out here; we’re still deciding on what form her content creation will take] We hope to bring our own personalities and perspectives on these issues, alongside humor, earnestness, and nuance. We want to help people make sense of the world we're in and think about what role they might play in the upcoming years of potentially rapid change. Our first long-form video For our first long-form video, we decided to explore AI Futures Project’s AI 2027 scenario (which has been widely discussed on the Forum). It combines quantitative forecasting and storytelling to depict a possible future that might include human extinction, or in a better outcome, “merely” an unprecedented concentration of power. Why? We wanted to start our new channel with a compelling story that viewers can sink their teeth into, and that a wide audience would have reason to watch, even if they don’t yet know who we are or trust our viewpoints yet. (We think a video about “Why AI might pose an existential risk”, for example, might depend more on pre-existing trust to succeed.) We also saw this as an opportunity to tell the world about the ideas and people that have for years been anticipating the progress and dangers of AI (that’s many of you!), and invite the br
 ·  · 3m read
 · 
Hi all, This is a one time cross-post from my substack. If you like it, you can subscribe to the substack at tobiasleenaert.substack.com. Thanks Gaslit by humanity After twenty-five years in the animal liberation movement, I’m still looking for ways to make people see. I’ve given countless talks, co-founded organizations, written numerous articles and cited hundreds of statistics to thousands of people. And yet, most days, I know none of this will do what I hope: open their eyes to the immensity of animal suffering. Sometimes I feel obsessed with finding the ultimate way to make people understand and care. This obsession is about stopping the horror, but it’s also about something else, something harder to put into words: sometimes the suffering feels so enormous that I start doubting my own perception - especially because others don’t seem to see it. It’s as if I am being gaslit by humanity, with its quiet, constant suggestion that I must be overreacting, because no one else seems alarmed. “I must be mad” Some quotes from the book The Lives of Animals, by South African writer and Nobel laureate J.M. Coetzee, may help illustrate this feeling. In his novella, Coetzee speaks through a female vegetarian protagonist named Elisabeth Costello. We see her wrestle with questions of suffering, guilt and responsibility. At one point, Elisabeth makes the following internal observation about her family’s consumption of animal products: “I seem to move around perfectly easily among people, to have perfectly normal relations with them. Is it possible, I ask myself, that all of them are participants in a crime of stupefying proportions? Am I fantasizing it all? I must be mad!” Elisabeth wonders: can something be a crime if billions are participating in it? She goes back and forth on this. On the one hand she can’t not see what she is seeing: “Yet every day I see the evidences. The very people I suspect produce the evidence, exhibit it, offer it to me. Corpses. Fragments of