> recherche-refl > the-bayesian-trap-veritasium

The Bayesian Trap

Veritasium - 2017-04-05

Bayes' theorem explained with examples and implications for life.
Check out Audible: http://ve42.co/audible
Support Veritasium on Patreon: http://ve42.co/patreon

I didn't say it explicitly in the video, but in my view the Bayesian trap is interpreting events that happen repeatedly as events that happen inevitably. They may be inevitable OR they may simply be the outcome of a series of steps, which likely depend on our behaviour. Yet our expectation of a certain outcome often leads us to behave just as we always have which only ensures that outcome. To escape the Bayesian trap, we must be willing to experiment.

Special thanks to Patreon supporters:
Tony Fadell, Jeff Straathof, Donal Botkin, Zach Mueller, Ron Neal, Nathan Hansen, Saeed Alghamdi

Useful references:
The Signal and the Noise, Nate Silver
The Theory That Would Not Die: How Bayes’ Rule Cracked the Enigma Code, Hunted Down Russian Submarines, and Emerged Triumphant from Two Centuries of Controversy, by Sharon Bertsch McGrayne

Bayes' theorem or rule (there are many different versions of the same concept) has fascinated me for a long time due to its uses both in mathematics and statistics, and to solve real world problems. Bayesian inference has been used to crack the Enigma Code and to filter spam email. Bayes has also been used to locate the wreckage from plane crashes deep beneath the sea.

Music from http://epidemicsound.com "Flourishing Views 3"

Pamphlet - 2017-04-05

"So what do you do for a living?"
"Oh the usual, drive to an open field, walk a mile, and talk to myself for a bit."

Saggitarius A - 2019-09-16

well that's one of the best thing YouTube offers

David Snyder - 2019-10-16

while ignoring the mosquito bites

Goedelite Kurt - 2019-11-11

Some people are unappreciative, perhaps because they are envious. They see the presenter of Veritasium as greatly privileged, making a living by strolling through a beautiful meadow and and able to think about mathematical ideas. They are underprivileged, such people, but not because they cannot stroll through a meadow for a living. Rather, they suffer from not being intellectually curious. They did not share the joy of discovery, as does Wordsworth's new child. Therefore, they are unappreciative.

Hugh Jones - 2019-11-26

"Fly my drone a bit, as well. Nice."

G dubb - 2019-12-28

@IndustrialDonut notice he changes hands quite often. Lol

ffttossenz - 2019-03-05

Normal probability YouTube tutorial: some dude in front of a white board
Veritasium: walking in nature talking about men in caves

Happy Raikhere - 2019-11-01

true af........ man

Shawn Noyes - 2018-04-30

Yes, this video inspired me to go back to school, and finish my degree. Class started today.
Thanks!
-Shawn

Leon - 2019-12-24

@sauce sauce lol. actually you'd need the data on the percentage of people who successfully graduate [insert program], and then separately the percentage of people who dropped out twice.

Tj - 2020-02-20

@Dennis Chen it's a joke, relax chan

d a t g i o - 2020-02-22

JUAN 12345 I mean I get that you’re just using common sense and you don’t know any batter, but many people even in their 50s have graduated with kids in elementary and everything.

Fontina Amoris - 2019-06-14

Interesting. I like how a mathematical theme evolved to become a philosophical one.

Greg Pearcey - 2019-11-09

math, like logic, is a language invented by humans to represent and communicate notions about a shared experience. Insofar as none of us are the repository of all of the available information, nor do we possess infallible processing power, any 'probability' we may ascribe to a notion is purely personal and subjective - i.e. philosophical.

Brian Lee - 2019-12-12

@WhimpyPatrol if only you could see my face after reading this

Darkwell0071 - 2019-12-26

Hayes theorem is all about philosophy.

Doug Thomson - 2019-12-31

Math and philosophy are inextricably linked.

Radosław Badurski - 2020-01-14

...You did hear about Vsauce, right?

Samuel Prevost - 2017-12-18

The turkey found that, on his first morning at the turkey farm, he was fed at 9 a.m. Being a good inductivist turkey he did not jump to conclusions. He waited until he collected a large number of observations that he was fed at 9 a.m. and made these observations under a wide range of circumstances, on Wednesdays, on Thursdays, on cold days, on warm days. Each day he added another observation statement to his list. Finally he was satisfied that he had collected a number of observation statements to inductively infer that “I am always fed at 9 a.m.”.
However on the morning of Christmas eve he was not fed but instead had his throat cut.

It doesn’t matter how many cases we list during our inductivist reasoning, nothing guarantees that the next case will lay in this inference we deducted from our observations, as the possible experiments and observations are infinite by number and type.
That's Russell's Inductivist Turkey

Hosein Ghanbari - 2019-10-30

Well, while this is a good example, I think this kinda completely goes with Bayes theory. How? Well, let's consider that turkey is more aware of the surroundings. That turkey is going with the wrong idea, while he (let's consider him a man!) is thinking about the probability of being fed, the real hypothesis can be the probability of being slaughtered. So the turkey goes with this idea that he his going to be fed on 9 am, which means it's less likely to be slaughtered at 9 am. In other words it depends on which variable or situation you are looking for. While the probability of being fed at 9 am is increasing, the probability of being slaughtered at 9 am is decreasing. But it doesn't mean it can't happen, "Everything is impossible until it's done". Also if we go for a correlation between the feeding and slaughter chance, maybe that poor turkey could live longer!
In other words I mean while the chance of being fed at 9 am was increasing, the chance of being slaughtered was also increasing! That's the problem when you approach a question with one assumption or ignoring other variables and outcomes! Lots of independent variables can lead to a result.

Shawn Willden - 2019-11-02

@Noah Hornberger If he were a really smart turkey, who understands Bayes' theorem and proper updating of beliefs based on evidence, he would say "It's highly probable that I am fed at 9, until the day I die".

registerme2 - 2019-11-24

The Turkey's logic was only incorrect from an external reference. As it is unlikely the Turkey ever realised that his logic was flawed.

Abhinav Chandraker - 2019-12-25

Ya. Similar to the Black Swan.

Bangla Desh - 2020-02-22

What is the probability of all turkeys that get fed at 9am having their throat cut

Caio Tavares - 2017-04-05

This 10 minute video was better at teaching Bayes Theorem than my whole Stochastic Processes class in the university.

Rasyid Al Faruqi - 2020-02-22

yes it does teach Bayes Theorem but not how to calculate it

Bangla Desh - 2020-02-22

Don't lie for internet points

Bangla Desh - 2020-02-22

@iona that's probably why they're not teaching high school

Bangla Desh - 2020-02-22

@orangeVikingfirefly you're comparing a university professor - someone who gives lectures over and over (multiple takes) - with someone who does not

orangeVikingfirefly - 2020-02-22

@Bangla Desh This guy does this for a living too, it's not his first video. Uni professors have to do it live, that is a huge difference, they can't stop and re-record everything (videos like this usually take tons of takes to make), they have to pay attention to a hundred other people while they do it, etc... And they only teach the same thing once a year, and usually they update the content, so it's not like they learn the thing by heart just because they teach it so much...

Jitesh Jangid - 2019-06-05

Man i think you should start a podcast.You talk really well.
Thank you.

NYB Creative Productions - 2019-06-21

100%

Kitten - 2019-08-09

@NYB Creative Productions Better not debate with you

Diego Marra - 2019-06-19

4:20 *Laughs in Heisenberg uncertainty principle*

mediawolf - 2019-05-26

I'm enjoying successively updating my understanding of Bayes' theorem by watching different people's takes on it. I also like that the moon is in this video.

megadeathx - 2019-06-14

That was bothering me. Is the moon at an orbit super far or is Venus super bright in the video pickup for whatever reason? Glad to have some added certainty to my assumed expectation. I am now 91% confident it's the moon.

sean h - 2019-07-18

"everything is impossible until it's done"
~ elizabeth holmes, theranos founder (hehehee)

The Souvlakinator - 2019-03-18

just like Ronnie Coleman said “If you always do what you’ve always done..you’ll always get what you always got”

HH Productions - 2019-01-08

0:00 Hey, this question is in my book of class 12 maths NCERT EXCERCISE 13.3 QUESTION 4,

Popcorn Sutton - 2019-08-10

@none none 2 blue balls, that teacher is fiiiine

Sumit Ray - 2019-08-16

Mine too

Y H - 2019-08-22

@Manisha Banerji Or better yet the complete collection of his writings as a free book on readthesequences.com

Manisha Banerji - 2019-08-22

@Y H And then become an AI safety philosopher

Siddhant Kaushik - 2019-11-26

Yes.
Indian stuff.

Nikolay Shiderov - 2019-06-01

I had heard about "Bayesian probability" and felt like learning more about it, as I know nothing about it.


The conclusion you draw at the end is about getting accustomed to results like rejection, which is something I have often dealt with and felt bad about. It gave me a new perspective on things.


I think you are an amazing speaker and have a great way of getting worthwhile ideas across. All the best.

Alonso S - 2019-05-30

I had never seen such a good explanation! I wish my professor was like this

Random Jin - 2019-01-03

If you want to get something you have never had before - start doing something you have never done before

Canes Venatici - 2019-09-14

Dislikes from 9% of people who actually got the disease after being diagnosed positive on the test.

Unwanted Commentary - 2019-06-07

Here I was thinking this would go into the Machine Learning uses, but instead you brought up some insightful philosophy. You rock!

Guillaume Laflèche - 2019-06-05

Thank goodness for thinkers, the world needs more of them.

CombraStudios - 2019-07-01

We know thinkers are useless when those who should listen don't

Cale Drey - 2019-08-28

4:28 "Made an analogy to a man coming out of a cave". Or should we say an allegory? #Plato

Raymond YN WONG - 2019-10-20

Exactly, and that's not by Price anyway. It's Plato

Michael Ibrahim - 2020-02-16

Raymond YN WONG Yes it is, Plato said something entirely different

John Edgewood - 2019-03-11

I say with 91% percent certainty that FLIES really like you in your video segments.

Charles Bukenya - 2019-05-05

Just beautifully said. You've earned a sub.

james caley - 2019-06-09

Otherwise known as the false positive paradox.

Isn't the Monty Hall problem an even better example of how maths can sometimes be a counter intuitive mind grenade.

bashrath1 - 2017-04-05

my favourite genre of trap

Jon Polley - 2018-04-29

KuraIthys tranny

Several Fighters - 2019-07-11

Admiral Ackbayes?

Matthew McKellar - 2019-07-18

@ozi Saying "No homo" is the most important step. At least that's what I've heard the other guy say...

The Dali Llama - 2019-10-05

I'm partial to honey traps.

tillyqtillyq - 2019-11-21

Wow this thread is some weird garbage

R Johnson - 2018-07-24

How I wish half my patients understood this.
Targeted testing.
That's the reason you go to an educated professional, instead of web-MD inc.

Please don't - 2019-11-19

8:04 And there's a saying "nothing is impossible", hence the Laplace smoothing.

T J - 2019-08-30

"Our actions play a role in determining outcomes and in determining how true things actually are" - well put.

Himanshu Bhoria - 2019-12-22

This question came in my exam yesterday. I had watched this video and I'm really thankful to you Derek <3

Manasi Snehal - 2019-09-26

Felt very nice when you asked at the end "is there anything like that you're thinking about? " :)

Werner Boden - 2018-07-17

Ok ... my previous statement in short:
The test is 99% accurate for postive and negative results.
You mix 2 different statistics together and derive false conclusions.

James Sibberman - 2019-06-02

Guess : 10%
Edit: nvm, close enough tho

notthere83 - 2019-05-12

"We should probably not have debates between people with 100% prior certainty and 0% prior certainty." - why does that remind me of today's political discourse...

The PC Security Channel [TPSC] - 2017-04-07

Great video. This is the kind of content I love to see. I don't care about the editing or production quality, as about the depth of content, the concept, and how well it is explained. Keep it up, Derek.

Arnold Christian - 2018-03-11

a comprehensive opinion and ideas which combine maths, physics, reality, philosophy, and essentially reflect upon the real world.

Grandfather_Din_Racket - 2018-05-18

+weallbfree Greetings, fellow libertarian...Bayes' utility, as it applies to libertarians, unites with libertarianism in the courts. For every bar-licensed prosecutor colluding with the bar-licensed judge to stack the jury against the defendant (who has a bar-licensed defense attorney), there is no counter-force. ...There was a tiny counterforce when I was handing out jury rights information, but since then, I haven't seen anyone do it properly (persuasively encouraging nullification, to the degree of comprehension encountered). For this reason, I suspect two things: 1--I will die soon. 2--When AGI is created, it will wipe the USSA and its inhabitants off the map as completely undeserving of both freedom and life. ...Of course, that's only if it cares about justice. If you donated to the Fedpath-controlled LP, it will put you in the group of people who could have exerted corrective control, but chose not to.

You'd better create a Libertarian Party if you want one to exist, because the few libertarians that are in the LP are in it to donate to an LNC comprised of (a single federal agent pushing the LP toward the Fedpath he desires)+(Several dilettante "anarchist" followers who have no idea how to pursue any viable strategy toward individual freedom who like wearing suits and pretending they're political radicals).

If you were a god-like AGI would you respect this? Would you respect "abolitionists" who didn't really care to abolish slavery?

I sure wouldn't.

Jonathan Jaget - 2018-05-24

Same here, I had to look away ...

Jon Doty - 2018-07-27

pierrecurie
I would never comment on the quality of a video, but I do agree. Keeping the camera still is paramount. I can focus on the content rather than the peripheral.

Koen Th - 2018-07-27

pierrecurie On the contrary, I was full of respect about the stability when I noticed the shadow: he was actually carrying the camera while filming himself walking. Not all too easy!

Cristian Pop - 2020-02-16

Aerial shot: one of the most beautiful places on earth
Handheld camera: DEATH BY FLIES!

ashfaque azad - 2019-05-30

Just read the book "The art of statistics" . It is brilliant.

David Wilkie - 2019-06-02

"Something like that.."
Everyone has a personal beginning, in a context of continuity, so it's a matter of circumstance what initial beliefs we have an innate trust in.
Every baby has a predisposition to be a tester of evidence (taster). When you are responsible for the baby, you want the best information available for looking after them.
That's why Science.

Avoidance Technologies - 2019-05-23

"If the test says it - it is probably true; however, some environments induce different results economically, and -> that's it".

A. N. - 2017-11-10

So if I procrastinate every day then it makes it even more likely that I procrastinate every day.

Very reassuring, thanks!

Smash Dreamworld - 2019-05-21

@Dave Messer Are you sure about that?

Daniel - 2019-07-03

the point is.... its never 100% certain, just more likely to be... but its not 100% true... so theres always a chance that ur not a procrastinator.

Annie Kallen - 2019-07-22

"The thing that we forget in Baye's theorem is that our actions play a role in determining outcomes and in determining how true things actually are." -- This video. Sorry, friend, you are not off the hook. ;)

AnteConfig - 2019-10-06

@A. N. Not necessarily. There might actually be a reason you're on a procrastination streak other than that you're a procrastinator.
Don't let yourself be fooled by illusions. Your beliefs, if nurtured, will come true.

Shawn Willden - 2019-11-02

@AnteConfig Your beliefs, if nurtured, may come true. Avoid 100% beliefs, because they cannot be updated by new evidence.

dddmemaybe - 2017-12-12

I like the thought "Set a Precedence" for this video.

A precedence for existences, to break illogical 100%'s and give way to constructiveness.
Let go of one hundred and find twenty-five fifties.

duhvn - 2019-09-12

The super skill here is hiking in a polo and not flinching at those bugs flying around your head. wow.

Mitten keat - 2018-02-25

Math and Philosophy while on a walk in Nature,
The way it's meant to be done.

towlie911 - 2018-12-06

I sleep too much, I feel like my life slips away to those morning hours. I can’t remember the last time I saw a sunrise. I should probably change that.

Marc Züst - 2017-04-05

1:49 It usually doesn't get emphasized enough how hard it is to get a good prior. For instance, Derek's prior of just taking the incidence in the general population is probably already invalid. Why? Well, you show the symptoms of the disease, that means you are already far more likely to have the disease than joe everyman next to you.
Why is this an issue? Well, Bayes' theorem is really good at showing you the highly counterintuitive odds of something being true, you just need to know the priors. And that is a huge wild card, everything stands and falls with the prior. You can use highly biased priors to "prove" your point, you just need to convince your audience that these priors are valid (and to some extent, Derek already did this by convincingly endorsing a likely inaccurate prior).

You could argue, for instance, that human made climate change is a myth even though we have strong evidence to suggest so. You could just say, well what are the odds that us tiny puny humans can have an impact on the global scale evironment? It's infinitesimal! Just look at the size of our planet! It's HUGE! So let's just input prior p=10^-13 or something (weight of all humans / weight of the earth). I know, it's ludicrous, but try arguing against that in front of a crowd that doesn't understand the intricate details of Bayes' theorem.

That notwithstanding, Bayes is great, and it is superior to standard frequentist statistics when used correctly. However, I feel the potential to "cheat with statistics" is also far more severe with Bayes. It doesn't solve all our problems.

tldr: Always check if the prior is valid!

Edit: spelling

Marc Züst - 2017-04-11

Thank you Michael, i got it, it's not that hard to understand. That was not my point, however. I simply wanted to talk about how hard it is to choose good priors, and that you can abuse Bayes by deliberately choosing priors that fit your needs. So I myself used an example to illustrate this point, just like Derek did. We're good right? Yeah we're good.

Michael Davey - 2017-04-11

Your example still doesn't demonstrate what you think it does. But, like you said, we're good.

CQUNC - 2017-04-20

"The specifics or plausibility of the example are completely irrelevant."

They are irrelevant to the logic, but adding such assumptions make the example harder to understand, which defeats its purpose.

Stephen Loftus-Mercer - 2017-05-08

If you plan on running multiple tests, it really doesn't matter what your first prior is, so long as it isn't 0% or 100%. Repeated applications of Bayes Theorem on successive inputs of data takes care of that. If you really don't know, picking 50% is always the right thing -- I have no data and I'm making a guess.

Guillaume Obozinski - 2019-11-01

Thank you for making this comment. Yes, in general if you get to do a medical test, using the frequency of the disease in the population is completely wrong ! Because the symptoms or even the circumstances that let you to see a doctor increase the prior odds significantly...

percheronphenom - 2019-06-15

So that was basically an understandable scientist doing a subtle life coaching on all of us who watched, while strolling up & down some nice landscape
I think we need more of that :) !!!

Marouf Bankousli - 2019-06-15

8:54 You broke my heart with those words..

Karan Bhatia - 2019-08-20

I can't believe how good the quality of his content is. Connectivity something what I learnt in school like Bayes theorem to reality and how it can change our perspective on real life #MindBlown

Edmond Beech - 2018-01-04

When you have a microeconomic exam in a few weeks and use this as revision

Ruiqiang Lin - 2017-04-06

Probability of getting a gf

Feralz - 2017-09-01

All in your head m8. Step outside your comfort zone. Have u tried she-males? #bayesian4life

Feralz - 2017-09-01

Yes the false positive is random, the total of false positives in relation to the whole population is random.

eyewars - 2018-12-05

0%

Kenivia - 2019-06-15

yes

GG 42 [GER] - 2018-06-26

9:27 when you have to hit that 500 word count in an exam

tomasgonzales - 2017-12-24

I remember this video came out the day before I was gonna get taught this at university lol, what are the chances.

Isaac Young - 2019-02-06

This is the epitome of what the "philosophy" is in a PhD or doctorate of philosophy...