Think more rationally with Bayes’ rule | Steven Pinker
Vložit
- čas přidán 4. 06. 2024
- The formula for rational thinking explained by Harvard professor Steven Pinker.
Subscribe to Big Think on CZcams ► / @bigthink
Up next, The war on rationality ► • The war on rationality...
In his explanation of Bayes' theorem, cognitive psychologist Steven Pinker highlights how this type of reasoning can help us determine the degree of belief we assign to a claim based on available evidence.
Bayes' theorem takes into account the prior probability of a claim, the likelihood of the evidence given the claim is true, and the commonness of the evidence regardless of the claim's truth.
While Bayes' theorem can be useful for making statistical predictions, Pinker cautions that it may not always be appropriate in situations where fairness and other moral considerations are important. Therefore, it's crucial to consider when Bayes' theorem is applicable and when it's not.
0:00 What is Bayesian thinking?
1:01 The formula
2:41 When Bayes’ theorem obscures the solution
4:25 Bayes’ theorem in a nutshell
Read the video transcript ► bigthink.com/series/explain-i...
----------------------------------------------------------------------------------
About Steven Pinker:
Steven Pinker is an experimental psychologist who conducts research in visual cognition, psycholinguistics, and social relations. He grew up in Montreal and earned his BA from McGill and his PhD from Harvard. Currently Johnstone Professor of Psychology at Harvard, he has also taught at Stanford and MIT. He has won numerous prizes for his research, his teaching, and his nine books, including The Language Instinct, How the Mind Works, The Blank Slate, The Better Angels of Our Nature, The Sense of Style, and Enlightenment Now: The Case for Reason, Science, Humanism, and Progress.
----------------------------------------------------------------------------------
Read more of our stories on rationality:
Bayesian search: A simple rule to find stuff you’ve lost
► bigthink.com/smart-skills/bay...
Hasty generalization: how to escape your biases and be more rational
► bigthink.com/thinking/hasty-g...
System 1 vs. System 2 thinking: Why it isn’t strategic to always be rational
► bigthink.com/the-well/system-...
----------------------------------------------------------------------------------
About Big Think | Smarter Faster™
► Big Think
The leading source of expert-driven, educational content. With thousands of videos, featuring experts ranging from Bill Clinton to Bill Nye, Big Think helps you get smarter, faster by exploring the big ideas and core skills that define knowledge in the 21st century.
► Big Think+
Make your business smarter, faster: bigthink.com/plus/
----------------------------------------------------------------------------------
Want more Big Think?
► Daily editorial features: bigthink.com/popular/
► Get the best of Big Think right to your inbox: bigthink.com/st/newsletter
► Facebook: bigth.ink/facebook
► Instagram: bigth.ink/Instagram
► Twitter: bigth.ink/twitter
What's an example of you can apply Bayes' Theorem?
The resurrection of Jesus Christ and the entire reliability of the Gospels, completely falls apart under Bayesian analysis.
@@jameshicks7125 I'm not sure I follow. Can you break it down for me? Thank you.
@@rainking50 likelihood = min, commonality of occurence = 0.
It is very interesting, and I am not sure how much it is Bayes' influence, that people don't even try but I think, a big part it is, for example when comparing rates of women in professor position in Universities in different countries, for example professors for Mathematics etc. so low in West-Europe and much higher in China or Russia or some East-Europe countries . I do not think that nowadays sexism in West-Europe can be spread in Unis at that mass, so apparently it is still Bayes' phenomenon 😅
Por qué no pusieron un ejemplo dentro del video?
Dr. Sagan: Extraordinary claims require extraordinary evidence.
Dr. Suess: Sometimes the questions are complicated and the answers are simple.
Occam's Razor: the simplest explanation that fits the evidence is often the best.
@@albarodriguez9406 Correction: The main gist of Occam's Razor is the most likely explanation is the one that makes the fewest number of assumptions.
@@scalbaldyfruub7499 this is true, sometimes complicated questions do have complicated answers, but so long as that complicated answer is based truly on things we KNOW and not just things we guess or assume, it's still the most likely answer
@@rosiepone Or put more simply,
"Extraordinary claims, require extraordinary evidence." As long as the evidence is there, good to go.
What in the ChatBot is this bullshit?!?!?
I'm an artist and I love this science/math/statistics stuff. I know too many artists who go by their feelings and never seem to look beyond themselves or established beliefs. And you get to interpret their work any way you like. The best novelists were the ones who actually did research and had their minds blown enough to want to write about their experiences. Bayes' rule applies to us artists just as much as it does to scientists. If we artists applied more scientific thought and processes to our art we would only created better work for everyone.
The scientist makes sense of the past, the artist is living and speaking to people in the future. Working together can be very productive in the present due to the way scientists become experts in their fields, and artists formulate ideas by linking expert thought. I think communication is part of the issue. Verbosity can be an enemy of the present
It's why I prefer having intellectual discussion with children
In spite of what you will read in ridiculous replies, Bayesian analysis, IS constantly used, not only in scientific evaluation, but also in art and creativity:
You CONSTANTLY update your views, vision, analyses.
THIS is Bayseain analysis, and HOPEFULLY it occurs constantly in artistic, physical, musical, and ANYthing one does in life.
Brains are Predictive organs, LONG evolved to use memory and EVERY novel sensory experience.
This includes your internal sensing of your own response, through the physical arousal, the temperature, the subtle hungers your system reports. This includes your relative energy, exhaustion, and all other interoceptive sensations.
Of course, THAT means that you CAN Misinterpret, as Pinker implies, things that are less or not associated, which you then correct through continuing Bayesian interpretation of what occurs in all systems - and everything is involved in systems -
though, as you see from"Alexbaddy" some systems, including "realizations" can be and ARE illusory. We see evidence that we are prepared to see, and lacunae, lacks, gaps, exist, arise, diminish, grow, disappear, in our lacunae, depending upon whether we are sufficiently critical of OUR OWN conclusions.
My wife heard the “horse vs. zebra” saying many times in the more than two decades it took before she was properly diagnosed. After seeing over 200 doctors, someone finally figured out that she was a zebra. Since then, she’s received proper care; but it was such a hard path to get there. I wish that more people had considered that zebras do exist.
Yeah it's a real problem, unfortunately tilted towards saving time and money on average rather than reaching the best outcome in each case.
Since she obviously visited experts I’m mostly surprised they didn’t notice the stripes.
Come on man marrying a Zebra is weird asf
@@ButterflyMatt They did. But they’d lose their jobs and be ostracized if they mentioned them.
Yeah. But the point is that nobody can live in a world where all the horses are treated as zebra’s. So although tough on your striped wife and you in this case, it’s best for the majority.
I think the transcription machinery wrote "Bayes rate" when Pinker actually said, "base rate" - a crucial distinction
Every time!
One of Profesor Pinnker's best qualities is his clarity of speech.
And his amazing hair 😎👍
@@eddieking2976 he is like A better looking, more refined, and modern beer mug
And his weird history with international pedophile kingpin Jeffery Epstein.
Think more rationally 👉The Connections (2021) [short documentary] 💖
Funny how you have a typo on the word speech 😅
A couple months ago I read 'Enlightenment Now!' by Prof. Pinker. It set me straight about several of my perceived notions with actual facts. It made me much more hopeful about humanity's future.
LMAO didnt you what happened durring the pandemic ? thats the true face of people
@@Chris-hw4mq Well, even so we got through it.
@@Chris-hw4mq Some did show their true face sure, but a huge percentage still isolated, wore masks, got vaccinated, and so on. And there are other places in the world than the US where things went a lot smoother.
@@Catlily5 of course we got through it, and nobody who did not get through it can tell us otherwise!
I'm not saying you shouldn't be hopeful, but Pinker is widely accused of naked data manipulation by social scientists.
Steven, wouldn't this video be more effective if you had included an example or two? I hope you will do so in a follow-on video.
Prior experience would seem to indicate that; certainly the evidence in teaching shows this is true...I expect the probability to be pretty high
Probably not his fault - it was likely just edited this way.
The challenge is most people on the planet are completely unaware of what is possible, so it is challenging to even identify the path to get there 4:28 That’s the equity. What do we need to study to move society as a whole along to some next higher level? Let’s learn that!
The issue with this mode of thinking is it tends to ignore outliers, as they are less likely to occur. So if you apply it across the board - as is done in the medical profession, hearing hooves and saying "horses" not "zebras", as Pinker said, you're going to miss the outliers. We could reach better outcomes by both considering the most likely cause, while simultaneously ruling out the dangerous outliers that while unlikely are still possible.
I agree. I believe that you could even give patients with anxiety and depression a great mental boost by assuring them they don't have some serious condition, if indeed they do not. But, from the perspective as an American here, much of our medical conduct these days here revolves more around money than helping people get diagnosed or assisted in going on with their lives normally and productively as they can. Once the conversation of your health goes to the money subject, the quality and thoroughness of your care has been sliced in half, at best. Sorry to go off on a tangent here but it is something that bothers me a lot. I think it also bothers some medical professionals, who seem almost guilty when they encounter somebody who cannot afford particular imaging or tests etc.
right , black swans walk on wall street
winning the lottery will happen
who wins is the outlier
At least my understanding of true Bayesian thinking would not only think what is most likely but leave open other possibilities rather than fixating on one.
That isn't a problem with bayes but rather the general lack of ability to think probabilistically. It stems from a general lack of education in the subject. The way he described here is more complicated than it has to be and it wasn't super complicated. There is a way it could be framed where even most children over say like 12yo could understand it. Most people are not very granular with probabilities, meaning they have a concept of something like 100%, 50/50 and 0% but the fact is that the space is continuous between 0 and 1 and there is no known limit to the number of decimal places. Any cut off point we pick is arbitrary. Just as an example, if I said something had a probability of 30% and it happened, most people would be surprised and would probably tell me afterwards "you were wrong!"
That's not a problem with Bayesian thinking though. You can use Bayesian thinking to guide which hypotheses you should test, incorporating both their probability and the consequences of if they're true.
Please post more of these by Pinker. He's great!
Think more rationally 👉The Connections (2021) [short documentary] 💖
My AB History major class teacher unknowingly taught us this, but in the context of scrutinizing a historical source
Great video. I have only one issue with it - what Steven Pinker said in the closing "If you could follow what I just said, you understand Bayes Theorem." While this is 100% true, viewers probably came to this video with this understanding.
Not necessarily - though it does seem that, because CZcams follows content algorithms, most people who saw this video were already predisposed to thinking this way.
Not as many people understand this principle as you might think, especially when they have a lifelong history of church indoctrination.
To be clear, I don’t think religion is inherently bad. I think it’s generally taught too young, and it’s taught in a very all-or-nothing way, as well as an overly literalist way.
Many Americans, if asked about the parable of loaves and fishes, would just say that literally happened and glean zero symbolic awareness from it - even though it’s a very clear symbol for the nature of generosity itself.
We don’t expect for food to multiply and feed thousands, but some things do grow when you share them. Goodness, empathy, the faith itself, compassion for the poor and marginalized… there are many ways to interpret what’s actually going on in the story.
But the least interesting one is also the one that indicates a high degree of literalism.
@@emilysmith2965 Trust me, I am in any way suggesting that everyone (or even a majority of people) understands Bayes Theorem. What I am saying is that only those who do would watch this video. I would venture to say that 90% of people never even heard of it and wouldn't bother clicking on it. I TOOK Probability in college (got an A) and had forgotten this was the name of the theorem.
In fact... you could use Bayes Theorem to calculate the probability that someone would watch the video. Sadly, this would come up quite low due to people's general tendency to avoid math altogether.
But, of those who do watch, the probability that the person already understands the topic is quite high.
Who would have thought a class 12 maths formula holds so much meaning. As someone said.... Maths and Nature are the Same. 😊
Or is it. The acuity of our perception of nature, and math's are often the same?
See Gödels Incompleteness Theorem
Think more rationally 👉The Connections (2021) [short documentary] 💖
This is a great series, I love it
This channel is amazing! And the Bayes' rule is incredible as well!
An adage doctors learned even in Medical schools is "common things occur commonly". It is the most succint intuitive expression of taking into account prior prevalence/incidence in Bayesian rule.
That's an interesting one. But it's also important to realize that rare events do happen sometimes. Earlier this year, my country saw a number of storms that were completely off the charts, completely unprecedented. A total WTF experience. It was incredible. You could almost say "there's a first time for everything" even if that's not quite true.
But even that can be Bayesian. Is it really just a “freak coincidence”? Or is global climate change also affecting the patterns of tropical storm seasons?
It stops being a “rare occurrence” to have the hottest summer on record if it’s BEEN the hottest summer on record for five years in a row.
Convergence to the norm.
A great book to read about this topic is "The Scout Mindset" by Julia Galef
Think more rationally 👉The Connections (2021) [short documentary] 💖
Thank you for explaining that!
Think more rationally 👉The Connections (2021) [short documentary] 💖
When I saw that movie, I felt like the boy in the story who travels to another world and then returns, only to find that it was all a dream, until he feels in his pocket and finds a handful of sand.
Which movie is it?
i wanna know what movie too!
Is it The Neverending Story?
Sounds a lot like Time Bandits. Great movie btw
Think more rationally 👉The Connections (2021) [short documentary] 💖
Occam's Razor is similar
it simply means you should compare the odds of the evdience with or without the context of hypotheis
The best Big Think videos are those, I think, that show the presenter or speaker continuously. I find the added video clips very distracting.
Yes and ruin the presentation.
Please don't spoil these great pieces by putting the music all over everything. It really detracts. Talking and informing is enough.
I'm thinking about how Occam's Razor fits into or fits with Bayesian thinking especially when it comes to hypothesis testing and theory formation.
One big problem is our tendency to infer causation from correlation. 'If I had not got stuck in traffic, I would have been on that plane that crashed, killing all on board. Who would have looked after my elderly mother then? God must have sent that traffic-jam to save me, because He has a purpose for me. It's a miracle!". As if none of the people that died had families and obligations of their own, and therefore made sure to leave enough time to get to the airport and catch their flight. Which is the more likely explanation: God punished the people who were more conscientious me because He loves my mother more than anyone else's; or I just caught some luck by screwing up? Is life sometimes unfair, or am I special?
I'm a firm Christian, however I think it's important for everyone to understand when you're crossing from observable rationale to accepting something by faith. If you get that confused, you'll potentially end up ignorant or disappointed.
Continue carefully updating your past presumptions.
belief is human feature. not everything can stop us making a belief. why do we ever limitting our best feature? we keep experimenting on it
Today is my maths exam and we have this formula in syllabus
The video's CC says "Bayes rate" several times when it should say "base rate".
Next time remember to mention E. T. Jaynes’ Max Ent - maximize entropy of the naive subjective prior. It will tank many iterations to wriggle out of the entropic corner of a lie.
Pinker neglects to dive deeper into the innumerable biases that misdirect our framing and analysis of the ‘Likelihood’, ‘Prior’ and ‘Evidence’ terms of the Posterior Probability equation.
Only so much you can do in 5 minutes.
can someone explain me what he said?? please
Wow!! Just realised that this is in CBSE's class 12th Maths book. But they never managed to explain it this well...
How do you side switch without gaslighting yourself into believing you're wrong? Or is that part of it?
It's was not enough to "Think Big" by showing the equation for Bayes' Rule. You needed to 'show small' by walking through at least one real world example. As complicated as the world we live in is surely there was one "If a train leaves a station..." example you could have plugged in to the equation at 1:11 that would have given it real meaning to the layman.
0:06 "Science popularizer" would be a good title for Kurzgesagt
Utilitarianism is also great in theory, as a formula for rationality may be, but what are the values and how do you quantify them?
But, if applied, the "Golden Rule" makes its value apparent.
"So in everything, do to others what you would have them do to you, for this sums up the Law and the Prophets."
"Give, and it will be given to you. A good measure, pressed down, shaken together and running over, will be poured into your lap. For with the measure you use, it will be measured to you.”
This is a side issue but one that is still Relevant. How much evidence is enough? How do you know that the next piece of the puzzle will not be more significant than the evidence so far. This gets even more of a problem when you are searching in your own memory or you ask a question like have I forgotten to consider some good evidence or argument that I just did not consider. When is enough enough!
Always +1 for Steve Pinker
Just not in this case, as he's wrong about it sometimes being desirable to not look at the base rate, such as when looking at crime rate.
Well, if you live in East Africa, the probability of it being a zebra is quite high. And as we get closer to the frontier of current knowledge, the probability of that something not even being an equine gets higher and higher while we still expect it to be some sort of a horse practicing our habitual Bayesian thinking.
You completely missed the point. If you're living in a place where zebras frequently roam, then the Bayesian prediction would be to assume it's a zebra and not a horse
@@TheCelticsAREboss Well, maybe I should've chosen another metaphor, but my point was that most of us live in a place with no zebras, but make general Bayesian assumptions about horses. And this rationale works just fine in normal conditions for most people, but fails for those who get closer to terra incognita of zebra-space where science gets really weird and commons sense is trying to fly out of the window.
@@TheCelticsAREboss I think he exactly got the point, and showed that hoof-sound example was flawed, or at least not fully explained.
@@2bfrank657 It's not flawed if you live anywhere where zebras don't exist, which includes the vast majority of humans on Earth.
OP was just quibbling about nonsense. I already explained how Bayesian's rule is applicable to an environment where zebra's frequent.
And there is nothing really to explain, at least not the audience that is going to be watching this video, i.e., people living in urbanization or the developed world.
Think more rationally 👉The Connections (2021) [short documentary] 💖
IMO, this badly needs a few concrete examples in conjunction with that formula.
The music is way too loud.
Sometimes the truth is hard to hear. So much so that we make it criminal to speak it.
This is why I am not religious. I feel it makes extraordinary claims without extraordinary evidence. Likelihood of someone coming back from the dead and changing water into wine has a base rate of 0 for me in my experience, so it’s nearly impossible for me to believe in unless I see evidence equally incredible as that. I feel the same is true for many things nowadays… people hear the slightest things and begin making assumptions off of evidence that could just as well mean nothing at all in the context of the topic. Next thing you know it’s being repeated on the news and spreads like fire.
I’m not saying I’m not open to believing in things, but I feel the evidence should be irrefutable. If everyone in the world understood the amount of work scientists have done to continuously question and improve their surroundings, maybe they wouldn’t be so quick to believe what they hear right away.
Sometimes the Humana's mind can't realize some evidence
Because he doesn't have that's ability.
@@ameermathkoor7113
It doesn’t have to, we have math and statistics for that.
I think you're not religious because you don't lack the strength others do, I'm not religious either but I wonder, should I have not been as strong mentally, would I have been desperate enough to turn to religion? I think many are forced to turn to religion because of their environment and circumstances, the mentally healthier you are, the better your living conditions in the early growing phase of life the less likely you're to ignore inner strength and the more likely you're to seek strength elsewhere
Religion is for the stronger people if we are being honest
to me mental strength is proportional to brain health, I wonder if there's research regarding brain health comparing those of religious and non-religious people
Describing what in effect is called "common sense" into a formula might be great, but, as he says around 3:15, you'd better know the limits of common sense, or you'd end up closing your eyes to the first scientific approach of simply Observing and Perceiving.
Overall I think this is a dangerous formula because there is not a side formula that calculates the limits of application of the Bayes formula.
The idea of Bayesian rationality takes me back years ago to Julia Galef.
0:01 who are the people in the stars ? Right side
1:43 so, the less evidence there is, the more probable it gets. Makes perfect sense.
I'm waiting the simple explanation of teh next theory, the one that evolved from this; free energy principal by Karl Frinston
I would write the equation like "post = prior x ( likely / evidence )", which hardly needs an explanation like that.
my teachers and poverty dissuded me to scince graduate so i gradate in commerce
Bayes theorem assumes IID variables (independent and identically disturbution), doest work when you're dealing with Moriarty Dr Holmes.
2:40
Light and L: *Bayesian strategizing activated*
Well said
Very informative
Posterior probability: 사후 확률, 경험적 확률
Credence: 신빙성
How much credence u had before u look at the evidence
Big fan of pinker.
It goes the other way round actually; if there is a lot of sexism in engineering [or surgeons etc] programs then the Bayes rate of women who want to be engineers will be pretty low.
prob. sexism given few women in eng. = (prob. few women in eng. given sexism X prob. sexism) / prob. few women in eng.
In my experience, sexism is pretty likely to keep women out of an industry if it is present.
IMO, there is a fair bit of sexism still around.
prob. there are few women in eng. is pretty much 1, at least for most fields of engineering.
What gets tricky is defining what level of sexism we talking about, and also, how little is "few"?
WOW ! I got it. The expectation value can be predetermined before you are handed evidence to the contrary !
I believe this boils down to "Feelings" vs. "Data" and whether they can be blended.
😂
Amazing video
Thank you ✨
So Bayes Therum comes off as more of a first approach but as part of a tool set. So basically its a balance of data history first, then real world data. Bayes = Prior Data + Current Data + Wisdom
i am here but did he just say women don't like being mechanical engingeers? given all the same opportunities, socially & economically, mechanical engineering is predisposed to attracting more men than women?
Does Bayes theorem predate Occam's razor?
Crazy to think we thought about everything we’re doing rn lol
me trying to figure out if my crush likes me back or not:
What a convoluted explanation. This guy is a professor?
I like your women in engineering analysis. The data indicates that women have a high probability of being people oriented and therefor less likely to be object oriented while males have a higher probability of being object oriented. We welcome women in engineering and when i had a few women in my class they were easier to manage, but women are less likely to WANT to be engineers.
Why are we easier to manage? Is it that we are focused on getting the job done rather than wasting time proving who has the biggest dick? Anyway, as a woman I wanted to get a well paid job so I became a chemical engineer. It’s an interesting and rewarding profession, but the money is the primary motivator. Do men want to become engineers for some other reason? I can’t imagine what it could be.
@@dianaf2077 The data indicates that men are more object oriented, so they tend to go into STEM jobs versus people oriented jobs. And for clarification over the span of 4 decades I do not recall any of my engineers attempting to prove genitalia size as they had there hands full with D.E.'s. And I am not aware of any male that would say females are easier to manage, as I was referring to the class management in general.
Umm…check what you wrote in your first post. You wrote “…when I had a few women in my class they were easier to manage.” And I wouldn’t mind if you gave a link to the source of this data that you are quoting.
@@dianaf2077 it is an incomplete sentence when you try to misconstrue it like you did. And to erase any ambiguity about my view, I preferred teaching when females where in the class. And if you are looking for a fight the data is from Dr. Jordan Peterson's research, so go fight with him if you are looking for some drama.
@@johnnychinstrap as an engineer, you will appreciate the important of precision in determining whether a structure a structured entity behaves as we intend it to; and from that point of view, I would suggest that your sentence is not fit for purpose, as it is not clear whether the pronoun 'they' refers to 'women', which was the subject of the preceding clause, or 'the class', as you intended. It certainly caused me to do a double-take.
Whilst I would agree that men and women have somewhat different cognitive aptitudes, they are not binary. We all have to deal with objects and we all have to deal with people, and aptitudes are a question of degree. Women are somewhat more agreeable than men, and men are somewhat inclined to obsess about the size of our equipment, but people confound these expectations all the time. However, humans, engineers and scientists included, have a tendency to make rules out of expectations, and science is littered with shameful examples of women whose interest in science was discouraged, whose abilities were not nurtured, and achievements not recognised, by the men in charge of the institutions to which they sought admittance. Rosalind Franklin and Jocelyn Bell, for example, were both overlooked for their contributions to work that won their male colleagues Nobel prizes. A century before they would not even have been allowed into the lab, or the lecture hall. Human potential cannot flourish without nurture and encouragement, seeds cannot grow if they are not planted. Science and engineering may be concerned with objective realities, but they are products of social collaboration, from which no-one should be excluded a priori. We should always remember that one thing that unites us all, whatever our differences, is human fallibility.
I’m struggling with the concept of a “forbidden” Bayes rate.
The closed captions made a mistake there: Prof. Pinker was saying "base rate", not "Bayes rate". It's an important concept, but not at all confined to Bayesian probability.
In his example of sexism in engineering programs, one critical base rate is the rate at which men and women inherently wish to become engineers. It is "forbidden" in some circles to suggest that this base rate is lower for women than men. This assumption naturally has a huge impact on the conditional probability that there is sexism in engineering programs, given the discrepancies between the genders.
Hi can anyone explain to me the part of about "10% female mechanical engineers". To conclude whether there is sexism or not, we need to see "how many females wanted to be mechanical engineers in the first place". Why is it so and how to apply Bayesian theorem? Your help is much appreciated!
Same reasoning can justify confirmation bias (prior "belief" supported by cherry-picking "evidence"). Pinker does suggest this is an issue.
Deal with your Bayes answer as a probability, not as a yes or no.
Can someone explain to me what they believe the differences would be between "Explain it Like I am Smart" and "Explain it Like I am Stupid"?
Sure. Explain it to me like I’m stupid it shorthand for I know nothing about this particular topic but I would like to know, so start off with the basics. Explain it to me like I’m smart is shorthand for I know a great deal about this particular topic and I have one question I would like to ask you but there is no need to start from first principles.
For example, if it is the first time I’m playing WOW I would ask you to explain it like I’m stupid. If I’m a neurosurgeon consulting another neurosurgeon about a patient I would say explain it to me like I’m smart.
‘Stupid’ and ‘smart’ are not being used in a judgmental or pejorative way, they’re being used as shorthand to let the explainer know your level of knowledge.
@@dianaf2077 Clear. Most appreciate your response and explanation. I was not exactly sure about the context.
Thrilled about the video but lol the formula notation is a little messed up. It should read as expected value of A given B = ... This reads as probability (or expected value) of A divided by B.
Interesting ☺️.
As someone with a rare illness, I also agree that doctors need to MORE OFTEN consider zebras.
The forward slashes used in this video are misleading. :)
and so we have the greatest reason for misunderstanding science...there is no ‘cause and effect’ in the natural world, only probabilities
such a clever breed you all are
Good content but the force feed background music is annoying especially when you listen as podcast
Intuition and common sense does the trick.
Every evaluation of evidence should always be considered in context
Not crime or educational stats - Huh, I wonder why 🤔?!
I do not get where applying Bayes is problematic. The examples show that applying Bayes is more fair. I do not get how fairness would be in contrast to Bayes inference, as mentioned. Bayes thinking is absolutely correct, wrong application of it is the problem, where people do not wish to consider less probable cases in tasks where it is their responsibility to examine all possible cases.
Thank you for giving us informative content. God bless you.
Love Steven Pinker
I definitely see a lot of the evidence
Thx Mr Pinker
sysetematic bakchodi around my life since childhood
Ha. And I always thought/heard it was/as "base theorem"
Basically don't rule put anything but consider the most like thing first
Hume taught us what it takes to be rational, Bayes taught us how to be rational
Annoying that the equation shown at 1:08 is typeset incorrectly. The left-hand side shouldn't be P(A/B), i.e. "probability of (A divided by B)", but P(A|B), i.e. "probability of A conditioned on B". Because A and B are meaningless to the reader, it would even better be stated as P(H|E), i.e. "probability of the hypothesis conditioned on the evidence", or "probability the hypothesis is true, given that we have observed the new evidence".
The same applies to the P(B/A) on the right side. Again, it should be P(B|A) or, better still, P(E|H).
Otherwise, not a bad exposition, although the digression into the base rate fallacy was a distraction and probably best dealt with separately.
Well, there's the hole : I can choose to use Bayes or not. Its a discretionary method .( I better not use the same discretion with gravity,especially if I am a mental health worker and I am confronted with a patient standing on the ledge outside my office window ,10 stories up) With Pinker ,we should be even more discretionary then he advises with Bayes: he's not worth the attention.
1) it is not clear how all of the mentioned items are quantified.
2) I think Steven Pinker is probably the only remaining credible public intellectual.
Dude Thats in My Maths Class (Probability)
Loving the cowboy boots too !
Anyone ever debate Pinker? I don't like his eloquent delivery of intentional misrepresentations and it would be nice to see a take down somewhere.
Basically Bayes takes the base rates into consideration and failing to do so is a logical fallacy called "base rate fallacy". Even if it is a fallacy there are times where those are useful. We evolved these cognitive biases for a reason. Wisdom requires understanding the limits of one's knowledge and that implies understanding that there is a difference between wisdom and knowledge. Generally speaking a key component in wisdom is context specific. Generally speaking it is irrational to not consider base rates but there are times where depending on the axioms, or the basic assumptions, that it is actually perfectly rational to ignore the base rates or to weight them. The reason is that we can't assume we have perfect information about base rates. As an example in the book "The Bell Curve" there is a controversial chapter which got the author labeled as a racist and while I guess that is a point of view a person could take, I personally looked at it as being something close to empirical proof that there was a systemic bias in terms of education resources available to certain minority groups for decades(several entire generations of people). If we were to look at that same data and first separate by economic status and then other metrics like race, nationality, or whatever, we would see that scores fall back in line like we would expect with a normal distribution. What that means in practical terms is that the majority of people, regardless of sex, race, or whatever, as an example, fall somewhere in the range of slightly below average intelligence to slightly above average intelligence. Things like races, sex, etc are independent of intelligence. What has a far larger impact and will actually move base rates around is economic status. Rich people get better education, on average, than poor people. Even though rich people are no more likely to be intelligent than a poor person they are more likely to be more educated and therefore score higher on just about all academic measures on average.
Someone talking sense on the internet? Now what are the odds that would have found something sensible when I clicked on this.
Nutritional claims. Like apple cider vinegar cures everything from hangnails to cancer for example. Josie Wales would ask, "How is it on stains?"
Distracting and unnecessary music. Captions show "Bayes rate" several times instead of "base rate".