back to indexJudea Pearl: Causal Reasoning, Counterfactuals, and the Path to AGI | Lex Fridman Podcast #56
link |
The following is a conversation with Judea Pearl,
link |
a professor at UCLA and a winner of the Touring Award
link |
that's generally recognized as the Nobel Prize of Computing.
link |
He's one of the seminal figures in the field of artificial intelligence,
link |
computer science, and statistics. He has developed and championed
link |
probabilistic approaches to AI, including Beijing networks,
link |
and profound ideas and causality in general. These ideas are important not
link |
just to AI, but to our understanding and practice of science.
link |
But in the field of AI, the idea of causality, cause and effect,
link |
to many, lie at the core of what is currently missing
link |
and will must be developed in order to build truly intelligent systems.
link |
For this reason, and many others, his work is worth returning to
link |
often. I recommend his most recent book called Book of Why that presents key
link |
ideas from a lifetime of work in a way that is accessible to the general public.
link |
This is the Artificial Intelligence Podcast. If you enjoy it,
link |
subscribe on YouTube, give it 5 stars on Apple Podcasts,
link |
support on Patreon, or simply connect with me on Twitter.
link |
Alex Friedman, spelled F R I D M A N. If you leave a review on Apple Podcasts
link |
especially, but also Cast Box or comment on YouTube,
link |
consider mentioning topics, people, ideas, questions, quotes, and science, tech, and
link |
philosophy that you find interesting, and I'll read them on this podcast.
link |
I won't call out names, but I love comments with kindness and thoughtfulness in them,
link |
so I thought I'd share them with you. Someone on YouTube highlighted a quote
link |
from the conversation with Noam Chomsky, where he said that the
link |
significance of your life is something you create. I like this line as well.
link |
On most days, the existentialist approach to life is one I find
link |
liberating and fulfilling. I recently started doing ads at the
link |
end of the introduction. I'll do one or two minutes after introducing
link |
the episode and never any ads in the middle that break the flow of the
link |
conversation. I hope that works for you and doesn't
link |
hurt the listening experience. This show is presented by Cash App,
link |
the number one finance app in the App Store. I personally use Cash App to
link |
have money to friends, but you can also use it to buy,
link |
sell, and deposit Bitcoin in just seconds. Cash App also has a new investing
link |
feature. You can buy fractions of a stock,
link |
say $1 worth, no matter what the stock price is.
link |
Brokerage services are provided by Cash App Investing, a subsidiary of Square,
link |
and member SIPC. I'm excited to be working with Cash App to
link |
support one of my favorite organizations called First, best known for their first
link |
robotics and Lego competitions. They educate and inspire hundreds of
link |
thousands of students in over 110 countries and have a
link |
perfect rating on charity navigator, which means the
link |
donated money is used to the maximum effectiveness.
link |
When you get Cash App from the App Store, Google Play,
link |
and use code LEX Podcast, you'll get $10 and Cash App will also donate $10
link |
to FIRST, which again is an organization that I've
link |
personally seen inspire girls and boys to dream of engineering a better world.
link |
And now, here's my conversation with Judea Pearl.
link |
You mentioned in an interview that science is not a collection of facts
link |
by constant human struggle with the mysteries of nature.
link |
What was the first mystery that you can recall that hooked you,
link |
that kept you in the creos... Oh, the first mystery, that's a good one.
link |
Yeah, I remember that. I had a fever for three days.
link |
When I learned about Descartes Analytical Geometry,
link |
and I found out that you can do all the construction in geometry using algebra,
link |
and I couldn't get over it. I simply couldn't get out of bed.
link |
What kind of world does Analytical Geometry unlock?
link |
Well, it connects algebra with geometry.
link |
Okay, so Descartes had the idea that geometrical construction
link |
and geometrical theorems and assumptions can be articulated in the language of
link |
algebra, which means that all the proof that we did in high
link |
school trying to prove that the three bisectors
link |
meet at one point and that, okay, all these can be proven
link |
by just shuffling around notation. Yeah, that was a traumatic experience.
link |
Traumatic experience. For me, it was. I'm telling you.
link |
It's the connection between the different mathematical disciplines that they all
link |
noted between two different languages.
link |
Languages. Yeah. So which mathematics discipline is most beautiful? Is
link |
geometry it for you? Both are beautiful. They have almost the same
link |
power. But there's a visual element to geometry
link |
being a visual. It's more transparent. But once you get over
link |
to algebra, then the linear equation is a straight
link |
line. This translation is easily absorbed.
link |
And to pass a tangent to a circle, you know, you have the basic theorems
link |
and you can do it with algebra. So but the transition from one to another
link |
was really, I thought that the card was the greatest
link |
mathematician of all times. So you have been at the, if you think of
link |
engineering and mathematics as a spectrum. Yes.
link |
You have been, you have walked casually along this spectrum
link |
throughout your, throughout your life. You know, a little bit of engineering and
link |
then, you know, a bit done a little bit of
link |
mathematics here and there. Not a little bit. I mean, we got a very solid
link |
background in mathematics because our teachers were geniuses.
link |
Our teachers came from Germany in the 1930s running away from Hitler.
link |
They left their careers in Heidelberg and Berlin and came to teach high school
link |
in Israel. And we were the beneficiary of that experiment.
link |
So I, and they taught us math the good way. What's a good way to teach math?
link |
Chronologically. The people. The people behind the theorems.
link |
Yeah. Their cousins and their nieces and their faces.
link |
And how they jumped from the bathtub when they screamed Eureka
link |
and ran naked in town. So you're almost educated as a historian of
link |
math. No, we just got a glimpse of that history
link |
together with the theorem. So every exercise in math was connected with the
link |
person and the time of the person. The period.
link |
The period also mathematically. Mathematically speaking, yes. Not the
link |
politics. No. So, and then in, in university,
link |
you have, you've gone on to do engineering. Yeah. I got a B.S. in engineering and a
link |
Technion. Right. And then I moved here for
link |
graduate work. And I got to, I did engineering
link |
in addition to physics in Radges. And it would combine very nicely with my
link |
thesis which I did in LCA laboratories and superconductivity.
link |
And then somehow thought to switch to almost computer science, software, even,
link |
even not switch but long to become, to get into
link |
software engineering a little bit. Yes.
link |
Programming, if you can call it that in the 70s. So there's all these disciplines.
link |
Yeah. If you were to pick a favor, what,
link |
in terms of engineering and mathematics, which path do you think
link |
has more beauty, which path has more power? It's hard to choose, no.
link |
I enjoy doing physics. I even have a vortex named on my name.
link |
So I have investment in immortality.
link |
So what is a vortex? Vortex is in superconductivity.
link |
In the superconductivity. You have permanent currents swirling around.
link |
One way or the other, you can have a store one or zero
link |
for a computer. That's what we worked on in the 1960s in LCA.
link |
And I discovered a few nice phenomena with the vortices.
link |
They pushed current and they moved. So that's a pearl vortex.
link |
Pearl vortex, right? You can google it. Right?
link |
I didn't know about it, but the physicists picked up on my thesis
link |
on my PhD thesis and it becomes popular. I mean thin film superconductors
link |
became important for high temperature superconductors.
link |
So they call it pearl vortex without my knowledge.
link |
I discovered it only about 15 years ago. You have footprints in all of the
link |
sciences. So let's talk about the universe a little bit.
link |
Is the universe at the lowest level deterministic or stochastic
link |
in your amateur philosophy view? Put another way, does God play dice?
link |
We know it is stochastic, right? Today. Today we think it is stochastic.
link |
Yes. We think because we have the Heisenberg
link |
uncertainty principle and we have some experiments to
link |
confirm that. All we have is experiments to confirm it.
link |
We don't understand why. Why is already...
link |
You wrote a book about why. Yeah, it's a puzzle.
link |
It's a puzzle that you have the dice flipping machine or God
link |
and the result of the flipping
link |
propagates with the speed faster than the speed of light.
link |
We can't explain it, okay? But it only governs microscopic phenomena.
link |
So you don't think of quantum mechanics as useful
link |
for understanding the nature of reality? No, diversionary.
link |
So in your thinking the world might as well be deterministic? The world is
link |
deterministic and as far as the new one firing is concerned
link |
it is deterministic to first approximation. What about free will?
link |
Free will is also a nice exercise. Free will is an illusion
link |
that we AI people are going to solve. So what do you think once we solve it that
link |
solution will look like? Once we put it in the page?
link |
First of all it will look like a machine.
link |
A machine that acts as though it has free will.
link |
It communicates with other machines as though they have
link |
free will and you wouldn't be able to tell the difference between
link |
a machine that does and machine that doesn't have free will.
link |
So the illusion, it propagates the illusion of free will amongst the other
link |
machines. And faking it is having it, okay?
link |
That's what Turing tells us all about. Faking intelligence is intelligent
link |
because it's not easy to fake. It's very hard to fake
link |
and you can only fake if you have it.
link |
Yeah, that's such a beautiful statement.
link |
Yeah, you can't fake it if you don't have it.
link |
So let's begin at the beginning with the probability
link |
both philosophically and mathematically. What does it mean to say
link |
the probability of something happening is
link |
50%. What is probability? It's a degree of uncertainty that an agent has
link |
about the world. You're still expressing some knowledge
link |
in that statement? Of course. The probability is 90%.
link |
It's absolutely different kind of knowledge and if it is 10%.
link |
But it's still not solid knowledge? It's solid knowledge.
link |
If you tell me that 90% assurance smoking will give you lung cancer
link |
in five years versus 10%, it's a piece of useful knowledge.
link |
So the statistical view of the universe, why is it useful?
link |
So we're swimming in complete uncertainty, most of everything.
link |
It allows you to predict things with a certain probability
link |
and computing those probabilities are very useful.
link |
That's the whole idea of prediction and you need prediction to be able to
link |
survive. If you can't predict the future then you're
link |
just crossing the street. It will be extremely
link |
fearful. And so you've done a lot of work in causation
link |
and so let's let's think about correlation. I started with the
link |
probability. You started with probability. You've
link |
invented the Bayesian networks and so we'll dance back and forth
link |
between these levels of uncertainty. But what is correlation?
link |
So probability is something happening, it's something,
link |
but then there's a bunch of things happening and sometimes they happen
link |
together, sometimes not, they're independent or not.
link |
So how do you think about correlation of things?
link |
Correlation occurs when two things vary together over a very long time.
link |
There's one way of measuring it or when you have a bunch of variables that
link |
they're all very cohesive. Then because we have a correlation here
link |
and usually when we think about correlation
link |
we really think causally. Things cannot be correlated unless there is a reason
link |
for them to vary together. Why should they vary together?
link |
If they don't see each other why should they vary together?
link |
So underlying it somewhere is causation. Yes.
link |
Hidden in our intuition there is a notion of causation because we cannot grasp
link |
any other logic except causation. And how does conditional probability
link |
differ from causation? So what is conditional probability?
link |
Conditional probability, how things vary, when one of them
link |
stays the same. Now staying the same means that I have chosen to look only
link |
of those incidents where the guy has the same value
link |
as previous one. It's my choice as an experimenter.
link |
So things that are not correlated before could become correlated.
link |
Like for instance if I have two coins which are uncorrelated
link |
and I choose only those flippings experiments in which a bell rings
link |
and a bell rings when at least one of them is a tail.
link |
Then suddenly I see correlation between the two coins
link |
because I only look at the cases where the bell rang.
link |
You see it's my design with my ignorance essentially
link |
with my audacity to ignore certain incidents.
link |
I suddenly create a correlation where it doesn't exist physically.
link |
Right so that's you just outlined one of the flaws
link |
of observing the world and trying to infer something
link |
from the math about the world from looking at the correlation.
link |
I don't look at it as a flaw. The world works like that.
link |
But the flaws come if we try to impose
link |
causal logic on correlation. It doesn't work too well.
link |
I mean but that's exactly what we do. That's what that has been the majority
link |
of science. The majority of naive science.
link |
So the stations know it. The stations know it if you
link |
condition on a third variable then you can destroy
link |
or create correlations among two other variables.
link |
They know it. It's in the data. It's nothing surprising.
link |
That's why they all dismiss the symptom paradox.
link |
Ah we know it. They don't know anything about it.
link |
Well there's there's disciplines like psychology where all the variables are
link |
hard to account for and so oftentimes there's a leap
link |
between correlation to causation. You're you're imposing a leap.
link |
Who is trying to get causation from correlation?
link |
Not you're not proving causation but you're sort of
link |
discussing it implying sort of hypothesizing without ability.
link |
Which discipline do you have in mind? I'll tell you if they are
link |
obsolete or if they are outdated or they are about to get outdated.
link |
Yes. Yeah tell me which one do you have in mind.
link |
Well psychology you know. It's okay what is it SEM?
link |
It's actually equation. No no I was thinking of applied psychology
link |
studying. For example we work with human behavior
link |
and semi autonomous vehicles. How people behave and you have to
link |
conduct these studies of people driving cars.
link |
Everything starts with the question. What is the research question?
link |
What is the research question? The research question. Do people fall asleep
link |
when the car is driving itself? Do they fall asleep or do they tend to fall
link |
asleep more frequently? More frequently. Then the car not driving.
link |
Not driving itself. That's a good question okay.
link |
And so you measure you put people in the car
link |
because it's real world. You can't conduct an experiment where you control
link |
everything. Why can't you? You could. Turn the
link |
automatic module on and off. Because it's on road public.
link |
I mean there's aspects to it that's unethical.
link |
Because it's testing on public roads. So you can only use vehicle.
link |
They have to the people the drivers themselves have to make that choice
link |
themselves. And so they regulate that. And so you just
link |
observe when they drive it autonomously and when they don't.
link |
And then. But maybe they turn it off when they were very tired.
link |
Yeah that's kind of thing. But you you don't know those variables.
link |
Okay so that you have now uncontrolled experiment.
link |
Uncontrolled experiment. We call it observational study.
link |
And we form the correlation detected. We have to infer causal relationship.
link |
Whether it was the automatic piece that caused them to fall asleep.
link |
Oh okay. So that is an issue that is about 120 years old.
link |
Yeah. I should only go 100 years old. Okay.
link |
And oh maybe it's no actually I should say it's 2000 years old.
link |
Because we have this experiment by Daniel.
link |
But the Babylonian king. That wanted the
link |
exile. The people from Israel that were taken
link |
in exile to Babylon to serve the king. He wanted to serve them king's food.
link |
Which was meat and Daniel as a good Jew couldn't
link |
eat a non kosher food. So he asked them to eat vegetarian food.
link |
But the king overseers says I'm sorry but if the king sees that your
link |
performance falls below that of other kids you know he's going to kill me.
link |
Daniel said let's make an experiment. Let's take four of us from Jerusalem.
link |
Okay give us vegetarian food. Let's take the other guys that
link |
to eat the king's food and in about a week's time
link |
we'll test our performance. And you know the answer.
link |
Of course he did the experiment. And they were
link |
so much better than the others. And the kings nominated them
link |
to super position in his king. So it was the first experiment.
link |
Yes. So there was a very simple it's also the same
link |
research questions. We want to know vegetarian food.
link |
Assist or obstruct your your mental ability.
link |
And okay so let's put the question is very old one.
link |
Even the democratic said if I could discover one cause
link |
of things I would rather discuss one cause and be a king of Persia.
link |
The task of discovering causes was in the mind of ancient people
link |
from many many years ago. But the mathematics of doing that
link |
was only developed in the 1920s. So science has left us often.
link |
Okay science has not provided us with the mathematics to capture the idea of
link |
x causes y and y does not cause x. Because all the questions of physics
link |
are symmetrical algebraic. The equality sign goes both ways.
link |
Okay let's look at machine learning. Machine learning today if you look at
link |
deep neural networks you can think of it as
link |
kind of conditional probability estimators.
link |
Beautiful. So where did you say that?
link |
Conditional probability estimators. None of the machine learning people
link |
clubbled you. Attacked you. Listen most people and this is
link |
why this today's conversation I think is interesting is most people would agree
link |
with you. There's certain aspects that are just
link |
effective today but we're going to hit a wall and there's a lot of
link |
ideas. I think you're very right that we're
link |
going to have to return to about causality.
link |
Let's try to explore it. Let's even take a step back.
link |
You've invented Bayesian networks
link |
that look awfully a lot like they express something like causation but they
link |
don't. Not necessarily. So how do we turn
link |
Bayesian networks into expressing causation? How do we build causal networks?
link |
This A causes B, B causes C. How do we start to infer that kind of
link |
thing? We start asking ourselves a question
link |
what are the factors that would determine the value of X? X could be
link |
blood pressure, death, hunger. But these are hypotheses
link |
that we propose. Hypothesis. Everything which has to do with
link |
causality comes from a theory.
link |
The difference is only how you interrogate the theory you have
link |
in your mind. So it still needs the human expert to propose.
link |
Right. You need the human expert to specify
link |
the initial model. Initial model could be very
link |
qualitative. Just who listens to whom? By whom listen to I mean
link |
one variable listen to the other. So I say okay the tide is listening to the
link |
moon and not to the rooster crow.
link |
And so far this is our understanding of the world in which we live.
link |
Scientific understanding of reality. We have to start there.
link |
Because if we don't know how to handle cause and effect
link |
relationship when we do have a model and we certainly do not know how to
link |
handle it when we don't have a model. So let's start first.
link |
In AI slogan is representation first, discovery second.
link |
But if I give you all the information that you need
link |
can you do anything useful with it? That is the first representation.
link |
How do you represent it? I give you all the knowledge in the world. How do you
link |
represent it? When you represent it I ask you
link |
can you infer x or y or z? Can you answer certain queries?
link |
Is it complex? Is it polynomial? All the computer science exercises we do
link |
once you give me a representation for my knowledge.
link |
Then you can ask me now I understand how to represent things. How do I discover
link |
them? At the second everything. So first of all I should echo the
link |
statement that mathematics and the current much of the machine learning
link |
world has not considered causation that A
link |
causes B just in anything. That seems like a
link |
non obvious thing that you think we would have
link |
really acknowledged it but we haven't. So we have to put that on the table.
link |
So knowledge how hard is it to create a knowledge from which
link |
to work? In certain area it's easy because we have
link |
only four or five major variables and an epidemiologist
link |
or an economist can put them down.
link |
Minimum wage, unemployment, policy, x, y, z and start collecting data
link |
and quantify the parameters that were left
link |
unquantified with the initial knowledge. That's the
link |
routine work that you find in experimental psychology,
link |
in economics, everywhere in the health science.
link |
That's a routine thing but I should emphasize you should start with a
link |
research question. What do you want to estimate?
link |
Once you have that you have to have a language of expressing what you want to
link |
estimate. You think it's easy? No. So we can talk about two things. I
link |
think one is how the science of
link |
causation is very useful for
link |
answering certain questions and then the other is how do we create
link |
intelligent systems that need to reason with causation.
link |
So if my research question is how do I pick up this water bottle
link |
from the table? All the knowledge is required to be able to
link |
do that. How do we construct that knowledge base?
link |
Do we return back to the problem that we didn't solve
link |
in the 80s with expert systems? Do we have to solve that problem
link |
of automated construction of knowledge?
link |
You're talking about the task of eliciting knowledge from an export.
link |
Task of eliciting knowledge from an export or the self discovery of
link |
more knowledge, more and more knowledge. So automating the building of
link |
knowledge as much as possible. It's a different game in the causal domain
link |
because it's essentially the same thing. You have to
link |
start with some knowledge and you're trying to
link |
enrich it. But you don't enrich it by asking for more rules. You enrich it by
link |
asking for the data, to look at the data and quantifying
link |
and ask queries that you couldn't answer when you started.
link |
You couldn't because the question is quite complex and it's not
link |
within the capability of ordinary cognition, of
link |
ordinary person or ordinary expert even to answer.
link |
So what kind of questions do you think we can start to answer?
link |
Even a simple one. Suppose I start with an easy one.
link |
What's the effect of a drug on recovery?
link |
What is the aspirin that caused my headache to
link |
be cured or what is the television program
link |
or the good news I received? This is already
link |
a difficult question because it's finding cause from effect.
link |
The easy one is find effects from cause. That's right.
link |
So first you construct a model saying that this is an important research
link |
question. This is an important question. Then you...
link |
I didn't construct a model yet. I just said it's an important question.
link |
And the first exercise is express it mathematically.
link |
What do you want to do? Like if I tell you what will be the effect
link |
of taking this drug? You have to say that in mathematics.
link |
How do you say that? Yes. Can you write down the question? Not the answer.
link |
I want to find the effect of the drug on my headache.
link |
Write it down. That's where the do calculus comes in.
link |
Yes. Do operator. What do you do operator? Do operator.
link |
Which is nice. It's the difference between association and intervention.
link |
Very beautifully sort of constructed. Yeah. So we have a do operator.
link |
So do calculus connected on the do operator itself connects
link |
the operation of doing to something that we can see.
link |
Right. So as opposed to the purely observing
link |
you're making the choice to change a variable. That's what it expresses.
link |
And then the way that we interpret it
link |
and the mechanism by which we take your query
link |
and we translate it into something that we can work with
link |
is by giving it semantics. Saying that you have a model of the world
link |
and you cut off all the incoming error into x
link |
and you're looking now in the modified mutilated model
link |
you ask for the probability of y. That is interpretation of doing x.
link |
Because by doing things you liberate them
link |
from all influences that acted upon them
link |
earlier and you subject them to the tyranny of your muscles.
link |
So you remove all the questions about causality by doing them.
link |
So there's one level of questions. Yeah.
link |
Answer questions about what will happen if you do things. If you do. If you drink
link |
the coffee or if you take the asthma. Right.
link |
So how do we get the doing data?
link |
Now the question is if we cannot run experiments.
link |
Right. Then we have to rely on observational study.
link |
So first we could start to interrupt. We could run an experiment
link |
where we do something. Where we drink the coffee and don't. And this
link |
the the do operator allows you to sort of be systematic about expressing.
link |
To imagine how the experiment will look like even though we cannot
link |
physically and technologically conduct it. I'll give you an example.
link |
What is the effect of blood pressure on mortality?
link |
I cannot go down into your vein and change your blood pressure.
link |
But I can ask the question. Which means I can even have a model of your body.
link |
I can imagine the effect of your how the blood pressure
link |
change will affect your mortality. How I go into the model and I conduct this
link |
surgery about the blood pressure. Even though
link |
physically I can do I cannot do it. Let me ask the quantum mechanics question.
link |
Does the doing change the observation? Meaning the surgery of changing the
link |
blood pressure is I mean no the surgery is
link |
very delicate. Very delicate. Incisive and delicate.
link |
Which means do x means I'm going to touch only x.
link |
Directly into x. So that means that I change only things which
link |
depends on x by virtue of x changing. But I don't depend
link |
things which are not depends on x. Like I wouldn't
link |
change your sex or your age. I just change your blood pressure.
link |
So in the case of blood pressure it may be difficult or impossible to construct
link |
such an experiment. No physically yes but hypothetically
link |
no. If we have a model that is what the model is for.
link |
So you conduct surgeries on a model you take it apart put it back
link |
that's the idea of a model. It's the idea of thinking counter factually
link |
imagining and that's the idea of creativity.
link |
So by constructing that model you can start to infer
link |
if the higher the blood pressure leads to mortality
link |
which increases or decreases. I construct the model I can still
link |
not answer it. I have to see if I have enough information in the model that
link |
would allow me to find out the effects of intervention
link |
from a noninterventional study from observation hands off study.
link |
So what's needed? We need to have assumptions about who
link |
affects whom. If the if the graph had a certain property
link |
the answer is yes you can get it from observational study.
link |
If the graph is too messy bushy bushy the answer is no you cannot.
link |
Then you need to find either different kind of observation that you haven't
link |
considered or one experiment. So basically does that
link |
that puts a lot of pressure on you to encode wisdom into that graph?
link |
Correct. But you don't have to encode
link |
more than what you know. God forbid if you put the like economists are doing that
link |
they call identifying assumptions they put assumptions even they don't
link |
prevail in the world they put assumptions so they can
link |
identify things. But the problem is yes beautifully put but the problem is you
link |
don't know what you don't know. So you know what you don't know because
link |
if you don't know you say it's possible it's possible
link |
that x affect the traffic tomorrow. It's possible you put down an error
link |
which says it's possible every error in the graph
link |
says it's possible. So there's not a significant cost to adding
link |
errors that the more error you add the better the less
link |
likely you are to identify things from purely
link |
observational data. So if the whole world is bushy
link |
and everybody affects everybody else
link |
the answer is you can answer it ahead of time
link |
I cannot answer my query from observational data I have to go to
link |
experiments. So you talk about machine learning
link |
is essentially learning by association or reasoning by association
link |
and this due calculus is allowing for intervention
link |
I like that word but action. So you also talk about counterfactuals
link |
and trying to sort of understand the difference in counterfactuals and
link |
intervention what's the first what is counterfactuals and
link |
why are they useful why are they especially useful as
link |
opposed to just reasoning what what affect actions have.
link |
What kind of factual contains what we normally call explanations.
link |
Can you give an example of a counterfactual? If I tell you that
link |
acting one way affects something I didn't explain anything yet
link |
but if I if I ask you was it the aspirin that cure my headache
link |
I'm asking for explanation what cure my headache
link |
and putting a finger on aspirin
link |
provide explanation it was aspirin it was responsible
link |
for your headache going away if if you didn't take the aspirin you would
link |
still have a headache. So by by saying if I didn't take
link |
aspirin I would have a headache you're thereby saying
link |
that aspirin is the thing that removes the headache.
link |
Yes but you have to have another important information
link |
I took the aspirin and my headache is gun
link |
it's very important information now I'm reasoning backward
link |
and I said what is the aspirin yeah by considering what would have happened
link |
if everything else is the same but I didn't take aspirin that's right so you
link |
know that things took place you know Joe killed Schmo
link |
and Schmo would be alive had John not used his gun
link |
okay so that is the counterfactual it had a conflict
link |
it has a conflict here or clash between
link |
observed fact that he he did shoot okay and the hypothetical
link |
predicate which says had he not shot you have a clash
link |
logical clash they cannot exist together that's the counterfactual and that
link |
is the source of our explanation of our the idea of responsibility
link |
regret and free will yeah so it certainly seems
link |
that's the highest level of reasoning right yes and philis is do it all the
link |
time who does it all the time physicist physicist
link |
in every equation of physics let's say you have a hook's law
link |
and you put one kilogram on the spring and the spring is
link |
one meter and you say had this weight been two
link |
kilogram the spring would have been twice as long
link |
it's no problem for physicists to say that
link |
accepted mathematics is only is in the form of equation
link |
okay equating the weight proportionality constant and the length of the
link |
string so you don't have the asymmetry in the
link |
equation of physics although every physicist things counterfactually
link |
ask high school kids had the weight been three kilograms
link |
what will be the length of the spring they can answer it immediately
link |
because they do the counterfactual processing in their mind
link |
and then they put it into equation algebraic equation
link |
and they solve it okay but the robot cannot do that
link |
how do you make a robot learn these relationships and why you would learn
link |
suppose you tell him can you do it so before you go learning
link |
yeah you have to ask yourself suppose that gives more information
link |
can the robot perform a task that i asked him to perform
link |
can he reason and say no it wasn't the aspirin it was the good news you
link |
received on the phone right because well
link |
unless the robot had a model a causal model of the world
link |
right right i'm sorry i have to linger on this but now we have to linger and we
link |
have to say how do we how do we do it how do we build
link |
yes how do we build a causal model without a team of human experts
link |
running around why don't you go to learning right away
link |
you're too much involved with learning because i like babies babies learn fast
link |
i forgot how they do it good yeah that's another question
link |
how do the babies come out with the counterfactual model of the world
link |
and babies do that yeah they know how to play with
link |
in the crib they know which balls hits another one
link |
and so they learn it by playful manipulation of the world
link |
yes there's a simple world involved only toys and balls and chimes
link |
but it's a if you think about it's a complex world
link |
we take for granted uh how complicated and kids do it by
link |
playful manipulation plus parent guidance pure wisdom and he'll say
link |
they meet each other can they say you you shouldn't have taken my toy
link |
right but and they these multiple sources of
link |
information they're able to integrate yeah so the challenge is about how to
link |
integrate how to form these causal
link |
relationship from different sources of data correct
link |
so how how how much is the information is it to play how much causal
link |
information is required to be able to play in the crib
link |
with different objects i don't know i haven't
link |
experimented with the crib okay not a crib picking up
link |
it's a very interesting manipulating physical objects on this very
link |
opening the pages of a book all the tasks the physical manipulation tasks
link |
do you have a sense because my sense is the world is extremely complicated
link |
it's extremely complicated i agree and i don't know how to organize it because
link |
i've been spoiled by easy problems such as
link |
cancer and death okay first we have to start
link |
no but it's easy the easy in the sense that you have only
link |
20 variables and they are just variables are not mechanics
link |
okay it's easy you just put them on the graph and they
link |
they speak to you yeah and you you're providing a methodology for
link |
letting them speak yeah i'm working only in the abstract
link |
the abstract knowledge in knowledge out data in between
link |
now can we take a leap to try and to learn
link |
in this very when it's not 20 variables but 20 million variables
link |
trying to learn causation in this world not learn but
link |
somehow construct models i mean it seems like you would only have
link |
be able to learn because constructing it manually
link |
would be too difficult do you have ideas of
link |
i think it's a matter of combining simple models
link |
for many many sources for many many disciplines
link |
and many metaphors metaphors are the basics of human intelligence basis
link |
yeah so how do you think of about a metaphor in terms of its use in human
link |
intelligence metaphors is an expert system
link |
an expert it's mapping problem
link |
with which you are not familiar to a problem with which you are familiar
link |
like i give you a good example the greek believed that the sky is an opaque
link |
shell it's not really a space infinite space
link |
it's an opaque shell and the stars are holes
link |
poked in the shells through which you see the eternal light
link |
it was a metaphor why because they understand how you poke holes in the
link |
shells okay they're not they were not familiar with
link |
infinite space okay and so and we are walking on a
link |
shell of a turtle and if you get too close to the edge you're gonna fall down
link |
to Hades or whatever yeah and that's a metaphor
link |
it's not true but this kind of metaphor enables Aristoteles
link |
to measure the radius of the earth because he said come on if we are
link |
walking on a turtle shell then a ray of light
link |
coming to this angle will be different this place
link |
will be different angle that's coming to this place i know the distance i'll
link |
measure the two angles and then i have the radius of the shell of the
link |
of the turtle okay and he did and he found his measurements
link |
were very close to the measurements we have today
link |
through the year what six thousand and seven hundred
link |
seven hundred kilometers of the earth that's something that would not occur
link |
to Babylonian astronomer even though the Babylonian experiments
link |
were the machine learning people of the time
link |
they fit curves and they could predict the
link |
eclipse of the moon much more accurately than the Greek
link |
because they fit curve okay that's a different metaphor
link |
something that you're familiar with a game a turtle shell okay
link |
what does it mean if you are familiar familiar means that answers to certain
link |
questions are explicit you don't have to derive them
link |
and they were made explicit because somewhere in the past
link |
you've constructed a model of that you're familiar with so the child is
link |
familiar with billiard balls yes so the child could predict that if
link |
you let loose of one ball the other one will bounce off
link |
these are you obtain that by familiarity familiarity is answering
link |
questions and you store the answer explicitly
link |
you don't have to derive them so this is idea for metaphor
link |
all our life all our intelligence is built around metaphors
link |
mapping from the unfamiliar to the familiar but
link |
the marriage between the two is a tough thing which i
link |
which we haven't been able to algorithmize so you think of that
link |
process of because of using metaphor to leap from one place to another
link |
we can call it reasoning is it a kind of reasoning
link |
it is reasoning by metaphor metaphor for metaphor do you
link |
think of that as learning so learning is a popular terminology today in a
link |
narrow sense it is it is it is definitely a
link |
so you may not okay right one of the most important
link |
learning taking something which theoretically is drivable
link |
and store it in accessible format i'll give an example chess
link |
okay finding winning winning starting moving chess
link |
is hard but uh it is there is an answer
link |
either there is a winning move for white or there isn't or there is a draw
link |
okay so it is the answer to that is available for the rule of the games
link |
but we don't know the answer so what does the chess master have that we don't
link |
have he has stored explicitly an evaluation of
link |
certain complex pattern of the board we don't have it
link |
ordinary people like me i don't know about you
link |
i'm not a chess master so for me i have to derive
link |
yes things that for him is explicit he has seen it before
link |
or you've seen the pattern before or similar pattern you see metaphor
link |
yeah and he generalized and said don't move is a dangerous move
link |
it's just that's not in the game of chess but in the game of
link |
billiard balls we humans are able to initially derive very effectively and
link |
then reasoned by metaphor very effectively and make it look so easy
link |
and it makes one wonder how hard is it to build it in a machine
link |
so in your sense how far away are we to be able to construct
link |
i don't know i'm not a futurist i can all i can tell you is
link |
that we are making tremendous progress in the causal reasoning
link |
a domain something that i even dare to call it revolution
link |
the causal revolution because what we have achieved in the past
link |
three decades is something that
link |
dwarf everything that was derived in the entire
link |
history so there's an excitement about current machine learning
link |
methodologies and there's really important good work you're doing
link |
in causal inference where do the word what is the future
link |
where do these worlds collide and what does that look like
link |
first they're gonna work without collisions
link |
it's gonna work in harmony harmony it's not the human is going to
link |
to jumpstart the exercise by providing qualitative
link |
noncommitting models of how the universe works
link |
how in reality the domain of discourse
link |
works the machine is going to take over from that point of view
link |
and derive whatever the calculus says can be derived
link |
namely quantitative answer to our questions
link |
these are complex questions i'll give you some example of complex question
link |
that will bugle your mind if you think about it
link |
you take result of studies in diverse population under diverse condition
link |
and you may infer the cause effect of a new population which doesn't even
link |
resemble any of the one studied and you do that
link |
by do calculus you do that by generalizing from one study to another
link |
see what's what's common with Beato what is different
link |
let's ignore the differences and pull out the commonality
link |
and you do it over maybe a hundred hospitals
link |
around the world from that you can get really
link |
mileage from big data it's not only you have many samples
link |
you have many sources of data so that that's a really powerful thing
link |
i think for especially for medical applications i mean
link |
cure cancer right that's how from data you can cure cancer
link |
so we're talking about causation which is the temporal
link |
temporal relationships between things not only temporal it was structural and
link |
temporal temporal enough temporal presence by
link |
itself cannot replace causation is temporal precedence
link |
the error of time in physics it's important it's necessary
link |
it's important yes is it yes i never seen cause
link |
propagate backward but if we call if we use the word
link |
cause but there's relationships that are timeless
link |
i suppose that's still forward in the era of time but
link |
the are there relationships logical relationships
link |
that fit into the structure
link |
sure the whole do calculate this logical relationship
link |
that doesn't require a temporal it has just a condition that
link |
it's you're not traveling back in time yes
link |
correct so it's really a generalization of
link |
a powerful generalization uh of what boolean logic yeah boolean logic
link |
yes that is sort of simply put and allows us to
link |
uh you know reason reason about the order of events the source the
link |
not about between we're not deriving the order of event
link |
we are given cause effects relationship okay
link |
they ought to be obeying the the time precedence relationship
link |
we are given that and now that we ask questions about
link |
other causal relationship that could be derived from the
link |
initial ones but were not given to us explicitly
link |
yeah like the case of the firing squad i give you
link |
in the first chapter and i ask what if rifleman a decline to shoot
link |
would the prisoners still be dead
link |
to decline to shoot it means that disobey order
link |
and the the rule of the games where that he is a
link |
obedient and marksman okay that's how you start that's the initial
link |
order but now you ask question about breaking the rules
link |
what if he decided not to pull the trigger he just became a pacifist
link |
and you can you and i can answer that the other rifleman would have
link |
killed him okay i want the machine to do that
link |
is it so hard to ask machine to do that it's just a simple task
link |
but if they have a calculus for that yes yeah
link |
but the curiosity the natural curiosity for me is
link |
that yes you're absolutely correct and important
link |
and it's hard to believe that we haven't done this
link |
seriously extensively already a long time ago so this
link |
this is really important work but i also want to know
link |
you know this maybe you can philosophize about how hard is it to learn
link |
okay let's assume a learning we want to learn it okay want to learn so what do
link |
we do we put a learning machine that watches execution
link |
trials in many countries and many
link |
locations okay all the machine can learn is to see
link |
shot or not shot dead not dead a court issued an order or didn't okay just the
link |
facts from the fact you don't know who listens to
link |
home you don't know that the condemned person
link |
listen to the bullets that the bullets are listening to the
link |
captain okay all we hear is one command two shots dead okay a
link |
triple of variable yes no yes no okay when that you can learn who
link |
listens to whom and you can answer the question no
link |
definitively no but don't you think you can start proposing ideas for humans to
link |
review you want machine to learn right you want a robot
link |
so robot is watching trials like that 200 trials
link |
and then he has to answer the question what if rifleman a
link |
refrain from shooting yeah so how to do that
link |
that's exactly my point at looking at the facts don't give you the strings
link |
behind the facts absolutely but do you think of machine learning
link |
as it's currently defined as only something that looks at the facts
link |
and tries right now they only look at the facts yeah so is there a way to
link |
modify yeah in your sense playful manipulation
link |
playful manipulation yes doing the interventionist kind of things
link |
intervention but it could be at random for instance the
link |
rifleman is sick that day or he just vomits or whatever so
link |
machine can observe this unexpected event which
link |
introduced noise the noise still have to be
link |
random to be able to and related to randomized experiment
link |
and then you have a observational studies from which to infer the strings
link |
behind the facts it's doable to a certain extent
link |
but now that we expert in what you can do
link |
once you have a model we can reason back and say what you kind of data you need
link |
to build a model got it so i know you're not a futurist
link |
but are you excited have you when you look back at your life
link |
long for the idea of creating a human level intelligence system yeah
link |
i'm driven by that all my life i'm driven just by one thing
link |
but i go slowly i go from what i know to the next step incrementally
link |
so without imagining what the end goal looks like do you imagine
link |
what the end goal is going to be a machine
link |
that can answer sophisticated questions counterfactuals of regret
link |
compassion responsibility and free will
link |
so what is a good test is a touring test
link |
a reasonable test free will doesn't exist yet
link |
there's no how would you test free will and that's so far we know only one
link |
thing many if robots can communicate
link |
with reward and punishment among themselves
link |
and hitting each other on the wrist and say you shouldn't have done that
link |
okay um playing better soccer because they can do that
link |
what do you mean because they can do that because they can communicate among
link |
themselves because of the communication they can do
link |
because they communicate like us reward and punishment
link |
yes you didn't pass the ball the right the right time
link |
and so for therefore you're going to sit on the bench for the next two
link |
if they start communicating like that the question is will they play better
link |
soccer as opposed to what as opposed to what they do now
link |
without this ability to reason about reward and punishment
link |
responsibility and I can only think about communication
link |
communication is and in not necessarily natural language but just
link |
communication just communication and that's important to have a quick
link |
and effective means of communicating knowledge
link |
if the coach tells you you should have passed the ball pink
link |
he conveys so much knowledge to you as opposed to what
link |
go down and change your software that's the alternative
link |
but the coach doesn't know your software so how can his coach tell you
link |
you should have passed the ball but that our language is very effective you
link |
should have passed the ball you know your software
link |
you tweak the right module okay and next time you don't do it
link |
now that's for playing soccer or the rules are well defined
link |
no they're not well defined when you should pass the ball is not well
link |
defined no it's a it's very soft very noisy
link |
yes you have to do it under pressure it's art
link |
but in terms of aligning values between computers and humans
link |
do you think this cause and effect type of thinking is important to align the
link |
values values morals ethics under which the
link |
machines make decisions is is the cause effect where
link |
the two can come together cause effect is necessary component
link |
to build a ethical machine because the machine has to empathize
link |
to understand what's good for you to build a model of your
link |
of you as a recipient which should be very much what what is compassion
link |
they imagine that you suffer pain as much as me as much as me
link |
i do have already a model of myself right so it's very easy for me to map
link |
you to mine i don't have to rebuild the model
link |
it's much easier to say oh you're like me okay therefore i would not hate you
link |
and the machine has to imagine it has to try to fake to be human essentially so
link |
you can imagine that you're that you're like me
link |
right and well who is me that's the fact that that's consciousness
link |
they have a model of yourself where do you get this model you look at yourself
link |
as if you are a part of the environment if you build a model of yourself
link |
versus the environment then you can say i need to have a model of myself
link |
i have abilities i have desires and so forth okay
link |
i have a blueprint of myself well not the full detail because i cannot
link |
get the halting problem but i have a blueprint
link |
so on that level of a blueprint i can modify things
link |
i can look at myself in the mirror and say if i change this more tweak this
link |
model i'm going to perform differently
link |
that is what we mean by free will and consciousness
link |
what do you think is consciousness uh is it simply self awareness so
link |
including yourself into the model of the world that's right that
link |
some people tell me no this is only part of consciousness and then they start
link |
telling me what they really mean by consciousness and i lose them
link |
yeah for me consciousness is having a blueprint of your software
link |
do you have concerns about the future of ai all the different trajectories of
link |
all of our research yes um where's your hope where the
link |
movement has where your concerns i'm concerned
link |
because i know we are building a new species
link |
that has the capability of exceeding our exceeding us
link |
exceeding our capabilities and can breathe itself
link |
and take over the world absolutely it's a new
link |
species that is uncontrolled we don't know the degree to which we control it
link |
we don't even understand what it means to be able to control this new
link |
species so i'm concerned i don't have anything to add
link |
to that because it's such a gray area that's unknown
link |
it never happened in history yeah the only the only time it happened in
link |
history was evolution with human being
link |
it wasn't very successful was it some people says it was a great success
link |
for us it was but a few people along the way
link |
our few creatures along the way would not agree
link |
so uh so it's just because it's such a gray area there's nothing else to say
link |
we have a sample of one sample of one it's us
link |
but some people would look at you and say yeah but we were looking to you
link |
to help us make sure that sample two works out okay actually we have more than
link |
a sample of more we have theories theories
link |
and that's good we don't need to be statisticians
link |
so sample of one doesn't mean any poverty of knowledge it's not
link |
sample of one plus theory conjectural theory
link |
of what could happen yeah that we do have
link |
but i i really feel helpless in contributing to this argument
link |
because i know so little and and my imagination is limited
link |
and i know how much i don't know and i but i'm concerned
link |
you're born and raised in israel born raised in israel yes and uh later
link |
served in um israel military defense forces in the in the israel defense force
link |
yeah what did you learn from that experience for this experience
link |
there's a kibbutz in there as well yes because i was in a nachal which is a
link |
combination of agricultural work and military service
link |
we were supposed i was really idealist i wanted to
link |
be a member of the kibbutz throughout my life
link |
and to live a communal life and uh so i prepared myself for that
link |
slowly slowly i want the greater challenge so that's a far world away
link |
both but i learned from that what i can it was a miracle
link |
it was a miracle that i served in the 1950s
link |
i i don't know how we survived the country was under austerity
link |
it tripled its population from 600 000 to a million point eight when i finished
link |
college no one who went hungry austerity yes
link |
when you wanted to buy uh to make an omelette
link |
in a restaurant you had to bring your own egg
link |
and the the imprisoned people from bringing food from the
link |
from farming from the villages to the city but no one went hungry and i
link |
always add to it and
link |
higher education did not suffer any budget cuts
link |
they still invested in me in my wife in our generation to get the best
link |
education that they could okay so i'm really
link |
grateful for the opportunity and i'm trying to pay back now okay
link |
it's a miracle that we survived the war of 1948
link |
they were so close to a second genocide it was all in plant
link |
but we survived it by miracle and then the second miracle that not many people
link |
talk about the next phase how no one went hungry
link |
and the country managed to triple its population
link |
you know it means to imagine united states going from what 350 million
link |
to yeah and believe this is a really
link |
tense part of the world it's a complicated part of the world
link |
israel and all around yes so religion is is um at the core of that complexity
link |
or one of the components religion is a strong motivating course to many many
link |
people in the middle east yes in your view
link |
looking back is religion good for society
link |
that's a good question for robotic you know
link |
there's echoes of that question equip robot with religious belief
link |
suppose we we find out or we agree that religion is good to you to keep you in
link |
in line okay should we give the robot the
link |
metaphor of a god as a matter of fact the robot will get it without us
link |
also why but the robot will reason by metaphor
link |
and what is the most primitive metaphor a child grows with
link |
mother smile father teaching father image and mother image that's god
link |
so you want it or not the robot will
link |
well assuming assuming that the robot is going to have a mother and a father
link |
it may only have a programmer which doesn't supply warmth
link |
and discipline discipline it does so the robot will have this a model of the
link |
trainer and everything that happens in the world cosmology and so is going to
link |
be mapped into the programmer that's god man
link |
the the thing that represents the origin of
link |
everything for that robot the most primitive relationship
link |
so it's going to arrive there by metaphor and so the the question is if
link |
overall that metaphor has served us well as humans I really don't know
link |
I think it did but as long as you keep in mind it's only a metaphor
link |
so if you think we can can we talk about your son
link |
yes yes can you tell his story a story well Daniel
link |
story is known is was abducted in Pakistan by al Qaeda driven sect and
link |
under various pretenses I don't even pay attention to what the pretence will
link |
originally they wanted to have to have in the United States
link |
deliver some promised airplanes there it was all made up and all this
link |
demands were bogus I don't know really but
link |
eventually he was executed in front of a camera
link |
at the core of that is hate and intolerance
link |
at the core yes absolutely yes we don't really appreciate
link |
the depth of the hate at which
link |
which billions of peoples are educated
link |
we don't understand it I just listened recently
link |
to what they teach you in Mogadishu
link |
when when the water stopped in the tap
link |
we knew exactly who did it the Jews the Jews we didn't know how
link |
but we knew who did it we don't appreciate what it means to us
link |
the depth is unbelievable do you think
link |
all of us are capable of evil
link |
and the education the indoctrination is really what we are capable of evil if
link |
you're indoctrinated sufficiently long and in depth
link |
we are capable of ISIS we are capable of Nazism
link |
yes we are but the question is whether we after we have
link |
gone through some Western education and we learned that everything is really
link |
relative it is no absolute God
link |
it's only a belief in God whether we are capable now
link |
of being transformed under certain circumstances
link |
to become brutal yeah that is a I'm worried about it
link |
because some people say yes given the right circumstances
link |
given economical bad economical crisis okay you are
link |
capable of doing it too and that's worries me I
link |
I want to believe it I'm not capable this seven years after Daniel's death he
link |
wrote an article at the Wall Street Journal titled Daniel
link |
Parle on the normalization of evil yes what was your message a message back then
link |
and how did it change today over over the years
link |
I I lost what was the message the message was that
link |
we are not treating terrorism as a taboo we are treating it as a bargaining
link |
device that is accepted people have grievance
link |
and they go and and bomb restaurants okay it's normal look you're even not
link |
not surprised when I tell you that 20 years ago you said what for
link |
grievance you go and blow a restaurant
link |
today it's becoming normalized the banalization of evil
link |
and we have created that to ourselves by normalizing by
link |
by making it part of
link |
political life it's a political debate every
link |
terrorist yesterday becomes a freedom fighter today and tomorrow it's becoming
link |
terrorist again it's switchable all right and so we should call out evil when
link |
there's evil if we don't want to be part of it
link |
becoming if you want yeah if we want to separate
link |
good from evil that's one of the first thing that
link |
what would they in the garden of Eden remember the first thing
link |
that god tell them was hey you want some knowledge here's a tree of good and evil
link |
so this evil touched your life personally
link |
does your heart have anger sadness or is it hope
link |
okay I see some beautiful people coming from
link |
Pakistan I see beautiful people everywhere but I see
link |
horrible propagation of evil in this country too
link |
it shows you how populistic slogans can catch the mind of the best
link |
intellectuals today's father's day I didn't know that yeah what's uh what's
link |
what's uh fond memory you have of Daniel
link |
or many good memories uh immense he was my mentor
link |
sense of balance that I didn't have
link |
yeah he saw the beauty in every person
link |
he was not as emotional as I am more
link |
looking things in perspective he really liked every person he really
link |
grew up with the idea that a foreigner is a reason for curiosity
link |
not for fear it's one time we went in Berkeley
link |
and homeless came out from some dark alley
link |
and said hey man can you spare a dime I retreated
link |
back you know two feet back and then I just hugged him and say here's a dime
link |
enjoy yourself maybe you want some
link |
money to take a bath or whatever where did he get it
link |
not for me do you have advice for young minds today dreaming
link |
about creating as you have dreamt creating
link |
intelligent systems what is the best way to arrive at new breakthrough
link |
ideas and carry them through the fire of criticism and
link |
and past conventional ideas ask your questions
link |
really your questions are never dumb
link |
and solve them your own way okay and don't take no for an answer
link |
look if they are really dumb you will find out quickly
link |
by trying an arrow to see that they're not leading any place
link |
but follow them and try to understand things your way
link |
that is my advice I don't know if it's going to help anyone
link |
now there's brilliantly there is a lot of
link |
it's the inertia in science in academia it is slowing down science
link |
yeah those two words your way that's a powerful thing
link |
it's against inertia potentially against the flow against your professor
link |
against your professor I wrote the book of why
link |
in order to democratize common sense
link |
in order to instill rebellious spirit in students so they wouldn't wait for
link |
until the professor gets things right
link |
as you wrote the manifesto of the rebellion against the professor
link |
against the professor yes so looking back at your life of research
link |
what ideas do you hope ripple through the next many decades
link |
what what do you hope your legacy will be
link |
I already have a tombstone
link |
carved oh boy and the fundamental law of
link |
counterfactuals that's what it's a it's a simple equation
link |
what a counterfactual in terms of a model surgery
link |
that's it because everything follows from that
link |
if you get that all the rest I can die in peace
link |
and my student can derive all my knowledge
link |
my mother mother means the rest follows yeah
link |
yeah thank you so much for talking today I really appreciate it thank you for
link |
being so attentive and instigating
link |
we did it we did the coffee helped thanks for listening to this conversation
link |
with jadea pearl and thank you to our presenting sponsor cash app
link |
download it use code lex podcast you'll get ten dollars
link |
and ten dollars will go to first a STEM education nonprofit
link |
that inspires hundreds of thousands of young minds
link |
to learn and to dream of engineering our future
link |
if you enjoy this podcast subscribe on youtube give it five stars on apple
link |
podcast support on patreon or simply connect with me
link |
on twitter and now let me leave you some words of wisdom
link |
from jadea pearl you cannot answer a question that you cannot ask
link |
and you cannot ask a question you have no words for
link |
thank you for listening and hope to see you next time