Lex Fridman Podcast - Judea Pearl: Causal Reasoning, Counterfactuals, Bayesian Networks, and the Path to AGI

Episode Date: December 11, 2019

Judea Pearl is a professor at UCLA and a winner of the Turing Award, that's generally recognized as the Nobel Prize of computing. He is one of the seminal figures in the field of artificial intelligence, computer science, and statistics. He has developed and championed probabilistic approaches to AI, including Bayesian Networks and profound ideas in causality in general. These ideas are important not just for AI, but to our understanding and practice of science. But in the field of AI, the idea of causality, cause and effect, to many, lies at the core of what is currently missing and what must be developed in order to build truly intelligent systems. For this reason, and many others, his work is worth returning to often. This conversation is part of the Artificial Intelligence podcast. If you would like to get more information about this podcast go to https://lexfridman.com/ai or connect with @lexfridman on Twitter, LinkedIn, Facebook, Medium, or YouTube where you can watch the video versions of these conversations. If you enjoy the podcast, please rate it 5 stars on Apple Podcasts or support it on Patreon. This episode is presented by Cash App. Download it (App Store, Google Play), use code "LexPodcast".  Here's the outline of the episode. On some podcast players you should be able to click the timestamp to jump to that time. 00:00 - Introduction 03:18 - Descartes and analytic geometry 06:25 - Good way to teach math 07:10 - From math to engineering 09:14 - Does God play dice? 10:47 - Free will 11:59 - Probability 22:21 - Machine learning 23:13 - Causal Networks 27:48 - Intelligent systems that reason with causation 29:29 - Do(x) operator 36:57 - Counterfactuals 44:12 - Reasoning by Metaphor 51:15 - Machine learning and causal reasoning 53:28 - Temporal aspect of causation 56:21 - Machine learning (continued) 59:15 - Human-level artificial intelligence 1:04:08 - Consciousness 1:04:31 - Concerns about AGI 1:09:53 - Religion and robotics 1:12:07 - Daniel Pearl 1:19:09 - Advice for students 1:21:00 - Legacy

The following is a conversation with Judea Pearl, a professor UCLA and a winner of the Turing Award that's generally recognized as the Nobel Prize of Computing. He's one of the seminal figures in the field of artificial intelligence, computer science, and statistics. He has developed and championed probabilistic approaches to AI, including Beijing Networks, and profound ideas in causality in general. These ideas are important not just to AI, but to our understanding and practice of science. But in the field of AI, the idea of causality, cause and effect, to many, lie at the core of what is currently missing and will must be developed in order to build truly intelligent

systems. For this reason, and many others, his work is worth returning to often. I recommend his most recent book, called Book of Why, that presents key ideas from a lifetime of work in a way that is accessible to the general public. This is the Artificial Intelligence Podcast. If you enjoy it, subscribe on YouTube, give it 5 stars in Apple Podcast, support it on Patreon, or simply connect with me on Twitter. Alex Friedman spelled F-R-I-D-M-A-N.

If you leave a review on Apple Podcast especially but also Cast Box or comment on YouTube, consider mentioning topics, people ideas, questions, quotes, and science, tech and and philosophy. You find interesting. And I'll read them on this podcast. I won't call out names, but I love comments with kindness and thoughtfulness in them. So I thought I'd share them with you. Someone on YouTube highlighted a quote from the conversation with No Chomsky, where he said that the significance of your life is something you create. I like this line as well. On most days, the existentialist approach to life is one I find liberating and fulfilling. I recently started doing ads at the end of the introduction. I'll do one or two minutes after introducing the episode and never any ads in the middle

that break the flow of the conversation. I hope that works for you and doesn't hurt the listening experience. This show is presented by CashApp. The number one finance app in the App Store. I personally use CashApp to sell money to friends, but you can also use it to buy, sell, and deposit Bitcoin in just seconds. CashApp also has a new investing feature. You can buy fractions of a stock, say $1 worth,

no matter what the stock price is. Broke your services are provided by Cashapp investing, a subsidiary of Square, a member of SIPC. I'm excited to be working with Cashapp to support one of my favorite organizations called the First, best known for their first robotics and legal competitions. They educate and inspire hundreds of thousands of students in over 110 countries and have a perfect rating on Charity Navigator, which means the donated money is used to the maximum effectiveness. When you get cash out from the App Store, Google Play, and use code Lex Podcast, you'll get $10 and cash out will also donate $10 to first, which again is

an organization that I've personally seen inspired girls and boys, the dream of engineering a better world. And now here's my conversation with you, Dea Pearl. You mentioned in an interview that science is not a collection of facts by constant human struggle with the mysteries of nature. What was the first mystery that you can recall that hooked you? That captor? Oh, the first mystery. That's a good one. Yeah, I remember that. I had a fever for three days when I learned about the carton, an allitered geometry. the card, an analytic geometry. And I found out that you can do all the construction in geometry using algebra. And I couldn't get over it. I simply couldn't get out of bed. What kind of world

does analytic geometry unlock? It connects algebra with the geometry. So the card had the idea that geometrical construction and geometrical theorems and assumptions can be articulated in the language of algebra, which means that all the proof that we did in high school and trying to prove that the three bisectos meet at one point and that okay all these can be proven by just shuffling around notation. Yeah that was a connection experience. The dramatic experience. For me it was, I'm telling you. So it's the connection between the different mathematical disciplines that they all know, different languages. Just languages. So which mathematics discipline is most beautiful?

Is geometry it for you? Both are beautiful. They have almost the same power. But there's a visual element, a geometry being of... Visual, it's more transparent. But once you get over to algebra, then the linear equation is a straight line. This translation is easily absorbed. And to pass a tangent to a circle, you have the basic theorems, and you can do it with algebra.

But the transition from one to another was really, I thought that the card was the greatest mathematician of all times. So you have been, if you think of engineering and mathematics as a spectrum, you have been, you have walked casually along this spectrum throughout your life, you know, a little bit of engineering and then, you know, I've done a little bit of mathematics here and there. But it's a bit, I mean, we got a very solid spectrum in mathematics because our teachers were geniuses. Our teachers came from Germany in the 1930s running away from Hitler. They left their careers in Heidelberg and Berlin and came to teach high school in Israel.

And we were the beneficiary of that experiment. And they taught us math a good way. What's a good way to teach math? Chronologically. The people. The people behind the exeorones, yeah. Their cousins and their nieces and their faces. And how they jumped from the bathtub when the scream Eureka and ran naked in town.

So you're almost educated as a historian of math? No, we just got a glimpse of that history together with the theorem. So every exercise in math was connected with the person. And the time of the person. The period. The period also mathematically speaking, yes, not the politics. No. And then in university, you have gone on to do engineering.

Yeah. I get to be as an engineering tech new one. And then I moved here for graduate work and I got to engineering in addition to physics in radgas. And it combined very nicely with my thesis which I did in LCA laboratories and superconductivity. And then somehow thought to switch to almost computer science software even even thought to switch to almost computer science software, even not switch, but long to get into software engineering a little bit, programming, if you can call that in the 70s. So there's all these disciplines. If you were to pick a favor, in terms of engineering and mathematics, which path do you think

has more beauty, which path has more power? It's how to choose, no. I enjoy doing physics and even have Votek's name on my name. So I have investment in immortality. So what is a Vortex? Votek's is in superconductivity. In the superconductivity? You have permanent comments swirling around.

One way or the other, you can have a store one or zero for a computer. That's what we worked on in the 1960s in HALCA. And I discovered a few nice phenomena with a vortices. So there's a problem. So there's a problem. So they move vortex. So you can google it. Right? I didn't know about it, but the physicist picked up on my thesis on my PhD thesis and it became popular. I mean, thin film superconductors

became important for high temperature superconductors. So they call it pale vortex without my knowledge. I discovered only about 15 years ago. You have footprints in all of the sciences. So let's talk about the universe a little bit. Is the universe at the lowest level deterministic or stochastic in your amateur philosophy view. Put another way, does God play dice? We know it is stochastic, right? Because today, today we think it is stochastic. Yes. We think because we have the Heisenberg Concert in the Principle, and we have some experiments

to confirm that. All we have is experiments to confirm it. We don't understand why. Why is always... You wrote a book about why. Yeah, it's a puzzle. It's a puzzle that you have a dice-safed flipping machine, a guard, and the result of the flipping propagated with the speed faster, the speed of light. We can't explain it, okay?

So it only governs microscopic phenomena. So you don't think of quantum mechanics as useful for understanding the nature of reality. No, they're visionally. So in your thinking, the world might as well be deterministic. The world is deterministic, and as far as the new one firing is concerned, it is deterministic to first approximation. What about free will? Free will is also a nice exercise.

Free will is a illusion that we AI people are going to solve. So what do you think once we solve it, that solution will look like? Once we put it in the page. If it doesn't look like, first of all, it will look like a machine. A machine that act as though it has free will. It communicates with other machines as though they have free will. And you wouldn't be able to tell the difference between a machine that does

and machine that doesn't have free will. So the illusion propagates the illusion of free will amongst the other machines. And faking it is having it. Okay, that's what two entities are about. Faking intelligence is intelligent because it's not easy to fake. It's very hard to fake. And you can only fake if you have it. That's such a beautiful statement.

Yeah, you could, yeah, you can fake it if you don't have it. So let's begin at the beginning with probability, both philosophically and mathematically, what does it mean to say the probability of something happening is 50 percent? What is probability? You disagree from certainty that an agent has about the world. You're still expressing some knowledge in that statement. Of course. The probability is 90% is absolutely different kind of knowledge. I mean, if it is 10%, but it's still not solid knowledge. It's still solid knowledge. But if you tell me that 90% assurance smoking will give you lung cancer in five years versus 10%. It's a piece of useful knowledge.

So the statistical view of the universe, why is it useful? So we're swimming in complete uncertainty. Most of everything allows you to predict things with a certain probability, and computing those probabilities are very useful. That's the whole idea of prediction. And you need prediction to be able to survive. If you can't predict the future, then you're just crossing the street will be extremely fearful. And so you've done a lot of work in causation.

And so let's think about correlation. I started with the probability. You started with probability. You've invented the Bayesian networks. Yeah. And so, you know, we'll dance back and forth between these levels of uncertainty. But what is correlation? What is it? back and forth between these levels of uncertainty.

But what is correlation? What is it? So probability of something happening is something, but then there's a bunch of things happening. And sometimes they happen together, sometimes not, they're independent or not. So how do you think about correlation of things? Correlation occurs when two things vary together

over a very long time, it's one way of measuring it, or when you have a bunch of variables that they are very aggressively, then we call it we have a correlation here. And usually, when we think about correlation, we really think cosy. Things are not becalling unless there is a reason for them to vary together. Why should they vary together? If they don't see each other, why should they vary together?

So underlying it somewhere is causation. Yes. Hidden in our intuition there is a notion of causation because we cannot grasp any other logic except causation because we cannot grasp any other logic except causation. And how does conditional probability differ from causation? So what is conditional probability? Conditional probability, how things vary when one of them stays the same. Now staying the same means that I have chosen to look only of those incidents

where the guy has the same value as previous one. It's my choice as an experimenter. So things that I'll not correlate it before could become correlated. Like for instance, if I have two coins which are uncorrelated, and I choose only those flippings experiments in which a bell rings, and a bell rings when at least one of them is a tail, then suddenly I see correlation between the two coins, because I only look at the cases for the bell rang. So you see, it's my design with my ignorance essentially, with my audacity to ignore certain incidents. I suddenly create a correlation where it doesn't exist physically. A correlation where it doesn't exist physically. Right, so you just outlined one of the flaws of observing the world and trying to infer

something from the matter about the world from looking at the correlation. I don't look at the flaws, the world works like that. If we try to impose causal logic on correlation, it doesn't work too well. I mean, but that's exactly what we do. That's what has been the majority of science. The majority of naive science. The decisions know it. The decisions know it. If you condition on a third variable, then you can destroy or create correlations among two other variables. They know it. It's in your data.

There's nothing surprising. That's why they all dismiss the Simpson-Barradok. Ah, we know it. It's not anything about it. Well, there's disciplines like psychology where all the variables are hard to account for and so Oftentimes there's a leap between correlation to causation. You're your leap Who is trying to get causation from correlation? Not you're not proving causation but you're sort of

discussing it, implying sort of hypothesizing with all the ability to. Which discipline we have in mind, I'll tell you if they are absolute, or if they are outdated, or they're about to get outdated. Yes. Oh, yes. Tell me which one do you have in mind? Oh, psychology, you know. It's okay, watch it, the SEM,

stuck to the grid. No, no, no, I was thinking of a plasticology studying. For example, we work with human behavior and semi-autonomous vehicles, how people behave. And you have to conduct these studies of people driving cars. Everything starts with a question. What is a research question? What is the research question?

What is the research question? The research question, do people follow sleep when the car is driving itself? Do they follow sleep or do they tend to follow sleep more frequently? More frequently. Then the car not driving itself. No, it's not driving itself. That's a good question, okay? And so you measure, you put people than the car not driving it's not driving it's a good question, okay?

And so you measure you put people in the car Because it's real world you can't conduct an experiment where you control everything Why can't you couldn't you could turn the Automotic module on and off Because it's Unroad public. I mean there's enough because it's on road public. I mean, there's aspects to it that's unethical because it's testing on public roads. So you can only use vehicle. They have to the drivers themselves have to make that choice themselves. And so they regulate that.

So you just observe when they drive it, the town of St. and when they don't. And then... But maybe they turn it off when they were very tired. Yeah, that's kind of thing. But you don't know those, they're... Okay, so that you have now uncontrolled experiments. Uncontrolled experiments.

We call it observational study. Yeah. And we form the correlation detected. We have to infer causal relationship. Whether it was the automatic piece has caused them to fall asleep. Oh, okay. So that is an issue that is about 120 years old. Yeah.

I should only go 100 years old. Okay. And oh, maybe it's no actually I should say is 2000 years old because we have this experiment by Daniel, but the Babylonian king that wanted the exile, the people from Israel that were taken in exile to Babylon, to serve the King. He wanted to serve them King's food, which was meat and Daniel as a good Jew couldn't eat a non-coach of food. So he asked them to eat vegetarian food, but the King overseer says, I'm sorry, but if the King sees that your performance falls below the head of other kids, you know, he's going to kill me. Daniel said, let's make an experiment. Let's take four of us from Jerusalem, okay, give us the Dittarian food.

Let's take the other guys to eat the King's food and about a week's time, we'll test our performance. And you know the answer, of course, he did the experiment and they were so much better than the others, and the King's nominated them to super position in his case. So it was a first experiment. So there was a very simple,

it's also the same research questions. We want to know a vegetarian food, assist or obstruct your mental ability. And okay, so the question is very old one. Even the democracy said, if I could discover one cause of things, I would rather discuss one cause than be a king of Persia. They, the task of discovering causes, was in the mind of ancient people, very many years ago. But the mathematics of doing this was only developed in the 1920s. So science has left us often.

Science has not provided us with the mathematics to capture the idea of X causes Y and Y does not cause X. Because all the equations of physics are symmetrical algebraic. The equality sign goes both ways. Okay, let's look at machine learning. Machine learning today, if you look at deep neural networks, you can think of it as kind of conditional probability estimators. Correct. Beautiful. So, where did you say that? Conditional probability estimators. None

of the machine learning people clouded you. I got you. Let's see. Most people, and this is why this today's conversation, I think, is interesting. Most people would agree with you. There are certain aspects that are just effective today, but we're going to hit a wall, and there's a lot of ideas. I think you're very right

that we're gonna have to return to about causality. And let's try to explore it. Let's even take a step back. You've invented Bayesian networks that look awfully a lot like they express something like causation, but they don't, not necessarily. So how do we turn Bayesian networks into expressive causation? How do we build causal networks? This a causes b because it's c how do we start to infer that kind of thing? We start asking ourselves a question. What are the factors that would determine the value of x?

that would determine the value of X. X could be blood pressure, death, hungry, hunger. But these are hypotheses that we propose. I hypothesis everything which has to do with causality comes from a theory. The difference is only what kind, how you interrogate the theory that you have in your mind. So it still needs the human expert to propose. Right.

You need the human expert to specify the initial model. Initial model could be very qualitative, just who listens to whom? By whom listen to I mean one variable listen to the other. So I say, okay, the tide is listening to the moon and not to the rooster crawl. And so forth, this is our understanding of the world in which we live. Scientific understanding of reality. We have to start there, because if we don't know how to handle cause-and-effect relationship, when we do have a model, and we certainly do not know how to handle it, when we don't have a model.

So let's start first. In AI slogan is representation first, discovery second. But if I give you all the information that you need, can you do anything useful with it? That is the first representation. How do you represent it? I give you all the knowledge in the world. How do you represent it? When you are represented, I ask you all the knowledge in the world, how do you represent it? When you are represented, I ask you, can you infer X or Y or Z?

Can you answer certain queries? Is it complex? Is it polynomial? Gold to computer science exercises. We do, once you give me a representation for my knowledge. Then you can ask me, now I understand how to represent things, how do I discover them? It's the second thing. So, first of all, I should echo the statement that mathematics and the current, much of the machine learning world has not considered causation that a causes b just in anything. So that that seems like us. That seems like a not obvious thing that you think we would have

really acknowledged it, but we haven't. So we have to put that on the table. So knowledge, how hard is it to create a knowledge from which to work? In certain areas it's easy because we have only four or five major variables in epidemiologists minimum wage, unemployment, policy, XYZ, and start collecting data and quantify the parameters that were left unquantified with initial knowledge. routine work that you find in experimental psychology, in economics, everywhere, in the health science, that's routine things. But I should emphasize, you should start with the research question, what you want to

estimate. Once you have that, you have to have a language of expressing what you want to estimate. You think it's easy? No. So we can talk about two things. I think one is how the science of causation is very useful for answering certain questions. And then the other is how do we create intelligence systems that need to reason with causation. So if my research question is how do I pick up this water bottle

from the table? All the knowledge is required to be able to do that. How do we construct that knowledge base? Does it, do we return back to the problem that we didn't solve in the 80s with expert systems? Do you have to solve that problem of automated construction of knowledge? Hmm. You're talking about the task of

at least listening knowledge from an expert. Task of eliciting knowledge from an expert, task of eliciting knowledge from an expert, or the self discovery of more knowledge, more and more knowledge. So automating the building of knowledge is much as possible. It's a different game in the causal domain, because essentially it's the same thing. You have to start with some knowledge and you're trying to enrich it. But you don't enrich it by asking for more rules.

You enrich it by asking for the data, to look at the data and quantifying and ask queries that you couldn't answer when you started. You couldn't because the question is quite complex and it's not within the capability of ordinary cognition, of ordinary person, or ordinary expert even, to answer. So what kind of questions do you think we can start to answer? Even simple, suppose, yeah, I start with easy one. What's the effect of a drug on recovery? What is the aspirin that caused my headache to be cured or what is the television program?

What the good notes I received? This is already a difficult question because it's fine the cause from effect. The easier one is find the effect from cause. That's right. So first you construct a model saying that this is an important research question. This is a point question. Then you do, I didn't construct a magic. I just said it's important question. A point question. And the first just said it's an important question.

And the first exercise is, express it mathematically. What do you want to do? Like, if I tell you, what will be the effect of taking this drug? You have to say that in mathematics. How do you say that? Can you write down the question? Not the answer. I want to find the effect of the drug on my headache.

Right. Write down. Write it down. That's where the dew calculus comes in. Yes. The dew operator. What do you do operator? The dew operator.

Yeah. That's just nice. It's the difference in association and intervention. Very beautifully sort of constructed. Yeah. So we have a dual operator. So dual calculus connected on the dual operator itself connects the operation of doing to something that we can see.

So as opposed to the purely observing, you're making the choice to change a variable. Let's put it in the it expresses. And then the way that we interpret it, and the mechanism by which we take your query and we translate it into something that we can work with, is by giving it semantics, saying that you have a model of the world

and you cut off all the incoming arrow into X and you're looking now in the modified mutilated model, you ask for the probability of Y. That is interpretation of doing X, because by doing things you liberate them from all influences that acted upon them earlier, and you subject them to the tyranny of your muscles. So you remove all the questions about causality by doing them. So you're not... There's one level of questions.

Answer questions about what will happen if you do things? If you drink the coffee, if you take the asthma. So how do we get the, once, how do we get the doing data? Now, the question is, if we cannot one experiment, right, then we have to rely on observational studies. So first we could decide to interrupt. We could run an experiment. Yeah.

Well, we do something where we drink the coffee and don't. And this, the, the do operator allows you to sort of be systematic about expressing it. Who, imagine how the experiment will look like, even though we cannot physically and technology conducted, I'll give you an example, what is the effect of blood pressure on mortality? I cannot go down into your vein and change your blood pressure,

but I can ask the question, which means I can even have a model of your body. I can imagine the effect of how the blood pressure change will affect your mortality. How? I go into the model and I conduct this surgery, about the blood pressure, even though physically I can do, I cannot do it. Let me ask the quantum mechanics question, does the doing change the observation? Meaning the surgery of changing the blood pressure is, I mean, no, the surgery is very delicate, it's very delicate, incinently delicate, which means, do means, do X means, I'm going to touch only X, only X. Directly into X. So that means that I change only things which depends on X,

by virtue of X changing. But I don't depend things which are not depends on X. Like I wouldn't change your sex or your age, I just change your blood pressure. So in case of blood pressure, it may be difficult to impossible to construct such an experiment. No, physically, yes. But hypothetically, no. If we have a model, that is what the model is for.

So you conduct surgeries on the model, you take it apart, put it back, that's the idea of a model. It's the idea of thinking about the fact that you're imagining and that's the idea of creativity. So by constructing that model, you can start to infer if the high blood pressure leads to mortality, which increases or decreases. By... I construct the model, I can still not answer it. I have to see if I have enough information in the model that would allow me to find out the effects of intervention from an non-interventional study, from a very hands-off study.

So what's needed to have assumptions about who affects whom. If the graph had a certain property, the answer is yes, you can get it from an observational study. If the graph is too meshy, bushy, bushy, the answer is no, you cannot. Then you need to find either different kind of observation that you haven't considered, or one experiment. So, basically, does that put... That puts a lot of pressure on you to encode wisdom into that graph, correct? But you don't have to encode more than what you know.

God forbid. If you put the like economists are doing that. They call identifying assumptions. They put assumptions, even if they don't prevail in the world, they put assumptions so they can identify things. But the problem is, yes, beautifully put. But the problem is you don't know, you don't know. So you know what you don't know. So you know what you don't know because if you don't know you say it's possible,

it's possible that X affect the traffic tomorrow. It's possible. You put down an arrow which says it's possible. Every arrow in the graph says it's possible. So there's not a significant cost to adding arrows that the more arrow you add, the less likely you are to identify things from purely observation and data. So if the whole world is bushy and everybody affects everybody else, the answer is you can answer it ahead of time. I cannot answer my query from observational data. I have to go to experiments. So you talk about machine learning is essentially learning by association or reasoning by association and this dueculus is allowing for intervention, I like that word, like that action. So you also talk about counterfactuals. And trying to sort of understand the difference in counterfactuals

and intervention, what is counterfactuals and why are they useful? Why are they especially useful as opposed to just reasoning what effect actions have? But kind of, the facturally contains what we normally call explanations. Can you give an example of that? If I tell you that acting one way affects something, I didn't explain anything yet. But if I ask you, was it the aspirin that killed my headache? I'm asking for explanation, what killed my headache? And putting a finger on aspirin,

provided that the financial, it was aspirin, it was responsible for your headache going away. If you didn't take the aspirin, you would still have a headache. So by saying, if I didn't take aspirin, I would have a headache. You're thereby saying that aspirin is the thing that removes the headache. He is, but you have to have another important information. I took the Asperin and my headache is gone. It's very important information. Now I'm reasoning backwards and I said, what is the Asperin?

Yeah. By considering what would have happened if everything else was the same, but I didn't take Asperin. That's right. So you know that things took place, you know? considering what would have happened if everything else was the same, but I didn't take Asperger. That's right. So you know, the things took place, you know, Joe killed Schmo. And Schmo would be alive.

Had John not used his gun. Okay, so that is the counterfactual. It had the conflict here, or clash between observed fact, that he did shoot, okay? And the hypothetical predicate, which says, had he not shot, you have a clash, a logical clash, they cannot exist together, that's a kind of fact. And that is the source of our explanation of our idea

of responsibility, regret and free will. Yes, it certainly seems that's the highest level of reasoning, right? Yes, and physicists do it all the time. Who does it all the time? Physicists. In every equation of physics, let's say you have a hook slow and you put one kilogram on the spring and the spring is one meter and you say, had this weight been two kilogram, the spring would have been twice as long. It's no problem for physicists to say that. Accepted mathematics is only for physicists to say that, except mathematics is only in the form of equation, equating the weight, proportionality constant, and the length of a string.

So you don't have the asymmetry in the equation of physics, although every physicist thinks Kant of actually, ask the high school kids, had the weight being three kilograms, what would be the length of the spring? They can answer it immediately, because they do the Kant of actual processing in their mind, and then they put it into a geobrari equation, and they solve it.

But the robot cannot do that. How do you make a robot learn these relationships? Why you put learn? Suppose you tell him, can you do it? Before you go learning, you have to ask yourself, suppose I give more information, can the robot perform a task, and I ask him to perform? Can reasonably say no? It wasn't the aspirin. It was a good news. You receive on the phone Right because well unless the robot had a model a causal

Model of the world Right, I'm sorry. I have to linger on this But now we have to linger and we have to say how do we do it? How do we build? Yes. How do we build the cause of model without a team of human experts? No, a book running around. Why don't you go to a learning right away?

You have too much involved with learning. Because I like babies, babies learn fast. I'm very worried about how they do it. Good. So that's another question. How do the babies come out with a counterfactual model of the world? And babies do that. They know how to play with the, in the crib.

They know which balls hits another one. And what they learn is by playful manipulation of the world. Yes. The simple world involved only toys and balls and chimes and bills. But it's ever, if you think about the complex world, we take for granted. Yes. How complex. And the kids do it by playful manipulation plus parents guidance, pure wisdom, and he'll say, they meet each other, can they say, you shouldn't have taken

my toy. And these multiple sources of information, they're able to integrate. So the challenge is about how to integrate, how to form these causal relationships from different sources of data. So how much is information is it to play? How much causal information is required to be able to play in the crib with different objects. I don't know. I haven't experimented with the crib. Okay, not a crib. I know it's a very interesting. Manipulating physical objects on this very opening the pages of a book, all the tasks, the physical manipulation tasks.

Do you have a sense? Because my sense is the world is extremely complicated. It's extremely complicated. I agree and I don't know how to organize it because I've been spoiled, but easy problems such as cancer and death. Okay. First we have to start trying. Easy, the easy sense that you have only 20 variables. Yes. And they are just variables, not mechanics. That's okay. Easy. You just put them on the graph and they speak to you.

Yes. And you're providing a methodology for letting them speak. Yes. I'm working only in the abstract. The abstract was knowledge in, knowledge out data in between. Now can we take a leap to try and to learn in this very, when it's not 20 variables, but 20 million variables, trying to learn causation in this world, not learn, but somehow construct models. I mean, it seems like you would only have to be able to learn because constructing it manually

would be too difficult. Do you have ideas of, I think it's a matter of combining simple models for many, many sources, for many, many disciplines, and many metaphors. Metaphors are the basics of human intelligence and basis. Yes, so how do you think of a bottom metaphor in terms of its use in human intelligence? Metaphors is an expert system. And experts, it's mapping problem with which you are not familiar to a problem with which you are familiar.

Like I give you a good example. The Greek believed that the sky is an opaque shell. It's not really an infinite space. It's an opaque shell and the stars are holes, where, in the shells through which you see the eternal light. That was a metaphor, why? Because they understand how you poke holes in shells. They were not familiar with infinite space.

They are not familiar with infinite space. And we are walking on a chair of a turtle. And if you get too close to the edge, you're going to fall down to hadith. There's a metaphor. It's not true. But this kind of metaphor enables Aristoteles to measure the radius of the earth, because he said, come on, if we are walking on a turtle shell, then the ray of light coming to this angle will be different, this place will be different angle that coming to this place. I know the distance, I'll measure the two angles, and then I have the radius of the shell of the turtle. And he did. And he found his measurement very close to the measurements we have today. It's through the year, about 6,700 kilometers there.

That something that would not occur to Babylonian astronomer, even though there are Babylonian experiments where the machine learning people of the time. They fit curves and they could predict the clips of the moon much more accurately than the Greek, because they fit curve. That's a different metaphor. Something that you're familiar with, a game, a total shield. What does it mean? It's your familiar. What does it mean? It's your familiar. Familiar means that answers to certain questions are explicit. You don't have to derive them.

And they were made explicit because somewhere in the past, you've constructed a model of that. You're familiar with the Chinese familiar with billion balls. So the child could predict that if you let loose of one ball, the other one will bounce off. You obtain that by familiarity. Familiarity is answering questions and you store the answer explicitly. You don't have to derive them. So this is ideal for metaphor. All our life, all our intelligence, is built around metaphors,

mapping from the unfamiliar to the familiar, but the marriage between the two is a tough thing, which we haven't yet been able to algorithmize. So you think of that process of, of using metaphor to leap from one place to another, we can call it reasoning. Is it a kind of reasoning? It is reasoning by metaphor, metaphor, metaphor. Do you think of that as learning? So learning is a popular terminology today in a narrow sense. It is, it is, it is, it is definitely a form. So you may not, okay, right? It's one of the most important learnings. Taking something which theoretically is drivable. But it is an answer. Either

there is a winning move for white or the isn't or there the easier draw. So the answer to that is available to the rule of the games. But we don't know the answer. So what does the chess master have? He has stored explicitly an evaluation of certain complex patterns of the board. We don't have it ordinary people like me, I don't know about you. I'm not a chess master. So for me I have to derive things that for him is explicit. He has seen it before, where he has and said, don't move the dangerous move. It's just that not in the game of chess, but in the game of

billionaire balls, we humans are able to initially derive very effectively and then reasoned by metaphor very effectively and make it look so easy that it makes one wonder how hard is it to build it in a machine. So in your sense how far away are we to be able to construct? I don't know, I'm not a futurist, I cannot I can tell you is it will be a making tremendous progress in the causal reasoning domain. Something that I even dare to call it evolution, the causal evolution, because what we have achieved in the past three decades is something that dwarf everything it was derived in the entire history. So there is an excitement about current machine learning methodologies

and there's really important good work you're doing in causal inference. good work you're doing in causal inference. Where do the, what is the future, where do these worlds collide, and what does that look like? First, they're going to work without collisions. It's going to work in harmony. Carbon is not the human is going to jump start. going to jump start the exercise by providing qualitative non-committing models of how the universe works, how the reality domain of discourse works. The machine is going to take over from that point of view and derive whatever

the calculus says can be derived, namely quantitative answer to our questions. These are complex questions. I give you some example of complex questions that bugle your mind if you think about it. You take result of studies in diverse population under diverse condition and you infer the cause of a new population which doesn't even resemble any of the ones studied. And you do that by do calculus. You do that by generalizing from one study to another. See, what's common with Plato? What is different? Let's ignore the differences and pull out the commonality.

And you do it over maybe a hundred hospitals around the world. From that, you can get really mileage from big data. It's not only do you have many samples, you have many sources of data. So that's a really powerful thing, I think, for, especially for medical applications, cure cancer, right? That's how from data, you can cure cancer. So we're talking about causation, which is the temporal temporal relationship between things. Not only temporal, it was structural and temporal. Temporal, enough, temporal presidents by itself cannot replace causation.

Is temporal precedence the era of time in physics? Yes, it's important, necessarily. Is it important? Yes. Is it? Yes, I've never seen cause propagate backward. But if we use the word cause, but there's relationships that are timeless. I suppose that's still forward and narrow of time. But are there relationships, logical relationships that fit into the structure? Sure, the whole, the whole, the whole, the logical relationship. That doesn't require a temporal. It has just the condition that it's you're not traveling back in time. Yes. Correct. So it's really a generalization of a powerful generalization of what? A Boolean logic. Yeah, Boolean logic. Yes. That is sort of simply put and allows us to

that is simply put and allows us to reason about the order of events, the source. Not about, between, we're not deriving the order of events. We are given cause of economic relationship. They ought to be obeying the time-president relationship. We are giving this. And now we ask questions about other cultural relationship that could be derived from the initial ones, but were not given to us explicitly. Like the case of the firing squad I give you in the first chapter. And I ask what if Rifleman A declined to shoot? Would the prisoners still be dead? To decline to shoot? That means

that he disobeyed order. And the rule of the games were that he is a obedient and marksman. That's how you start. That's the initial order. But now you ask questions about breaking the rules. What if he decided not to pull the trigger? He just became a pacifist. And you can answer that. The other life, for example, would have killed him. I want the machine to do that. It's so hard to ask the machine to do that. It's such a simple data.

But they have a calculus for that. But the curiosity, the natural curiosity for me is that, yes, you're absolutely correct and important. And it's hard to believe that we haven't done this seriously, extensively, already a long time ago. So this is really important work. But I also want to know, you know, this, maybe you can philosophize about how hard is it to learn? Okay, let's just learn.

We want to learn it, okay? We want to learn. So what do we do? We put a learning machine that watches execution trials in many countries and many locations, okay? All the machine can learn is to see, shot or not shot, dead, not dead. A code issued in order or didn't, okay? That's the fact. From the fact you don't know who listens to whom. You don't know that the condemned person, listen to the bullets, that the bullets are listening to the captain, okay? All we hear is one command, two shots dead, okay? A triple of variables. Yes, no, yes, no. Okay. When that you can learn who listens to whom and you can answer the question, no.

Definitely no, but don't you think you can start proposing ideas for humans to review? You want machine to deny it, right? You want a robot. So robot is watching the trials like that, 200 trials and then he has to answer the question, what if rifleman A refrained from shooting? How do you do that? That's exactly my point. Looking at the facts, don't give you the strings behind the facts. Absolutely. But do you think of machine learning as this currently defined as only something that looks at the facts and tries to do it? Right now they only look at the facts. So is there a way to modify in your sense? Playful manipulation. Playful manipulation. Yes. Do it interventionist kind of things. Intervention.

It could be a random, for instance, the rifleman is sick the day when he just vomits. So whatever. So we can observe this unexpected event, which introduced noise. The noise still have to be random, to be able to relate it to randomize experiment. And then you have a observational studies from which to infer the strings behind effects.

It's doable, to a certain extent. But now that we are expert in what you can do once you have a model we can reason back and say what you kind of data you need to build a model. Got it. So, I know you're not a futurist but are you excited? Have you, when you look back at your life, long for the idea of creating a human level intelligence system? Yeah, I'm driven by that. All my life, I'm driven just by one thing. But I go slowly. I go from what I know to the next step incrementally.

So without imagining what the end goal looks like. Do you imagine what an end goal is going to be a machine that can answer sophisticated questions, counterfactuals of regret, compassion, responsibility, and free will. So what is a good test? Is it touring tests? It's a reasonable test. It's a free will, doesn't exist yet. There's no... How would you test free will? That's... It's a reasonable task. A reasonable task. A reasonable task. A reasonable task.

A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task.

A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task. A reasonable task.

A reasonable task. A reasonable task. How would you test, free will? That's so far, we know only one thing. I mean, if robots can communicate with reward and punishment, and hitting each other on the wrist and say you shouldn't have done that. Okay. Playing better soccer because they can do that. What do you mean because they can do that? Because they can communicate among themselves.

Because of the communication they can do the soccer. Because they communicate like us. Reward and punishment. Yes, you didn't pass the ball the right time and so forth They are for you're gonna sit on the bench for the next two If they start communicating like that the question is will they play better suck up as opposed to what as a port what they do now without this ability to reason about Reward and punishment responsibility and reward and punishment, responsibility. And...

In fact, I can only think about communication. Communication is not a so natural language, but just communication. Just communication. And that's important to have a quick and effective means of communicating knowledge. If the coach tells you you should have passed the ball, pink, he conveys so much knowledge to you as opposed to what? Go down and change your software. That's the alternative. But the coach does know your software.

So how can the coach tell you you should have passed the ball? But our language is very effective, you should have passed the ball. You know your software, you tweak the right module. Okay? And next time you don't do it. Now that's for playing soccer or the rules are well defined. No, no, no, not well defined. When you should pass the ball, the ball doesn't. No, it's very soft, very noisy. You, you have to do that with pressure. It's art. But in terms of aligning values between computers and humans,

do you think this cause and effect type of thinking is important to align the values, values, morals, ethics under which the machines make decisions is, is the cause effect where the two can come together. Because the fact is necessary component, to build a ethical machine, because the machine has to empathize, to understand what's good for you, to build a model of you, the recipient, which should be very much, what is compassion? They imagine it you suffer pain as much as me. As much as me.

I do have already a model of myself. So it's very easy for me to map you to mine. I don't have to rebuild a model. It's much easier to say, oh, you're like me. Okay, therefore I would not hate you. And the machine has to imagine, has to try to fake to be human. Essentially, so you can imagine that you're like me, right? And what over who is me? That's the further, that consciousness.

They have a model of yourself. Where do you get this model? You look at yourself as if you are a part of the environment. If you build a model of yourself versus the environment, then you can say, I need to have a model of myself. I have abilities, I have desires and so on. I have a blueprint of myself, not a full detail, because I cannot get the holding problem, but I have a blueprint. So on that level of a blueprint, I can modify things. I can look at myself in the mirror and say, hmm, if I change this tweet this mile, I'm going to perform differently. That is what we mean by free will. And consciousness. What do you think is consciousness? Is it simply self-awareness, including yourself into the model of the world?

That's right. Some people tell me, no, this is only part of consciousness. And then they start telling me, what do you really mean by consciousness? And I lose them. For me, consciousness is having a blueprint of your softwa. Do you have concerns about the future of AI, all the different trajectories of all of our research? Yes. Where's your hope, where the movement has, where your concerns? I'm concerned because I know we are building new species that has the capability of exceeding us, exceeding our capabilities

and can breathe itself and take over the world absolutely. It's a new species, it is uncontrolled. We don't know the degree to which we control it, we don't even understand what it means to be able to control these new species. So I'm concerned. I don't have anything to add to that because it's such a grey area, that unknown. It never happened in history. The only time it happened in history

was evolution with human being. It wasn't very successful, wasn't it? Some people say it was a great success. For us it was, but a few people on the way, or a few creatures along the way would not agree so So it's just because it's such a great area. There's nothing else to say We have a sample of one sample of one. It's us

But we don't want people to look at you and say Yeah, but we were looking to you to help us make sure that the sample too works out okay. Actually, we have more than a sample of more. We have theory of theories. Yeah. And that's a good. We don't need to be statisticians. So a sample of one doesn't mean a pavarti of knowledge. It's not. A people say, one plus theory, conjectural theory, of what could happen. That we do have. But I really feel helpless in contributing to this argument, because I know so little,

and my imagination is limited. And I know how much I don't know. And I'm concerned. You're born and raised in Israel. Born and raised in Israel, yes. And later served in the military defense forces. In the Israel defense force. Yeah What did you learn from that experience? There's a kibbutz in there as well. Yes, because I was in the nachar which is a

Compination of agricultural work and military service and a combination of agricultural work and military service. We were supposed, I was really idealist. I wanted to be a member of the Kibbutz throughout my life and to leave a communal life. And so I prepared myself for that. Slowly, slowly, I went to the greater challenge. Prepared myself for that Slowly slowly I Wanted to greet the challenge So that's a there's a far world away both what I learned from that what I can either

It was a miracle It was a miracle that I served in the 1950s I don't know how we survived. The country was under austerity. It tripled its population from 600,000 to a million point eight when I finished college. No one went hungry. Osterity yes. When you wanted to buy, to make an omelette, in the restaurant you had to bring your own egg. And the... the imprisoned people from bringing food from the farm,

from farming, from the villages to the city. But no one went hungry. And I always add to it, and higher education did not suffer any budget cuts. They still invested in me, in my wife, in our generation generation to get the best education that they could. Okay. So I'm really grateful for the opportunity. And I'm trying to pay back now. Okay. It's a miracle that we survived the war of 1948. They were so close to a second genocide. It was all planned.

But we survived it by a miracle, and then the second miracle that not many people talk about, the next phase. How no one went hungry in the country managed to triple its population. You know what it means to triple the to imagine United States going from what 350 million to a million. Yeah, and believe that's a really tense part of the world. It's a complicated part of the world. Israel and all around. Religion is at the core of that complexity, one of the components. The religion is a strong motivating course to many, many people in the Middle East. Yes.

In your view, looking back, is religion good for society? That's a good question for robotic, you know? I call it a quick robot with religious beliefs. That's a good question for robotic, you know. I cause that question. We probode with religious beliefs. Suppose we find out who we agree that religion is good to keep you in line. Should be given a robot in the metaphor of a guard. The metaphor of a robot will get it without us also. Why? The robot will reason by metaphor.

And what is the most primitive metaphor? A child grows with. Mother, smile, father, teaching, father image, and mother, that's God. So, what do you want it or not? The robot will... Assuming that the robot is going to have a mother and a father, it may only have a programmer, which doesn't supply warmth and discipline. What discipline it does, so the robot will have a model of the train hall.

And everything that happens in the world, cosmology and so it's going to be mapped into the programmer. It's God. Man. The thing that represents the origin of everything for that robot. It's the most primitive relationship. So it's going to arrive there by metaphor. And so the question is if overall that metaphor has served as well as humans. I really don't

know. I think it did. But as long as you keep in mind, it's only metaphor. So if you think we can, can we talk about your son? Yes, yes. Can you tell his story? A story? Daniel. A story is known. He was abducted in Pakistan by al-Qaeda-driven sect, and under various pretenses, I don't even pay attention to what the pretence will. Originally they wanted to have a United States deliver some promised airplanes. It was all made up, all this demands were bogus.

I don't know really, but eventually he was executed in front of a camera. At the core of that is hate and intolerance. Is the core? Yes is hate and intolerance. At the core? Yes, absolutely, yes. We don't really appreciate the depth of the hate, which billions of people are educated. We don't understand it. I just listen to listening to what they teach you in Mogadishu. When the water stops in the tap, we knew exactly who did it, the Jews.

We didn't know how, but we knew who did it. We don't appreciate what it means to us. The depth is unbelievable. Do you think all of us are capable of evil? And the education, the indocrination is really what creates the world. Absolutely, yeah, capable of evil. If you are indoctrinated sufficiently long and in depth, you are capable of ISIS, you

are capable of Nazism. Yes, we are. But the question is whether we have to, we have gone through some Western education and we learned that everything is really relative. It is no absolute God. It is only a belief in God. Whether we are capable now of being transformed under formed under certain circumstances to become brutal. Yeah, that is a, I'm worried about it because some people say, yes, given the right circumstances, given economical, bad economical crisis, you are capable of doing it too. That's always me. I want to believe it. I'm not capable. This is seven years after Daniel's death. He wrote an article at the Wall Street Journal titled

Daniel Parallel and Normalization of Evil. Yes. What was your message back then and how did it change today, over the years? I lost. What was the message? The message was that we are not treating terrorism as a taboo. We are treating it as a bargaining device that is accepted. People have grievance and they go and bomb restaurants, okay? It's normal. Look, you're even not not surprised when I tell you that. 20 years ago you say, what? For grievance, you go and blow a restaurant. today is become normalized, the banalization of evil. And we have created

that to ourselves by normalizing, by making it part of political life. it's a political debate. Every terrorist yesterday becomes a freedom fighter today and tomorrow it becomes terrorist again, it is fritiable. And so we should call out evil when there is evil. If we don't want to be part of it, we want to separate good from evil. That's one of the first things that in the Garden of Eden, remember the first thing that God tells him, he wants some knowledge.

Here is the tree of good and evil. So this evil touched your life personally. Does your heart have anger, sadness or is it hope? I see some beautiful people coming from Pakistan. I see beautiful people everywhere, but I see a horrible propagation of evil in this country too. It shows you how populistic slogans can catch the mind of the best intellectuals. Today is Father Day?

I didn't know that. Yeah, I heard it. What's a fond memory you have of Daniel? Oh, many good memories. He means he was my mentor. Brilliant. He had a sense of balance that I didn't have. Yeah.

He saw the beauty in every person. He was not as emotional as I am, more look-and-things in perspective. He really liked every person. He really grew up with the idea that a foreigner is a reason for curiosity, not for fear. At one time we went in Berkeley and homeless came out from some dark alley and said, Hey man, can you spare a dime? I would reach it back, you know, two feet back. And then he just hugged him and say, here's a dime, enjoy yourself, maybe you want some money to take a bus or whatever.

Where did he get it? Not for me. Do you have advice for young minds today, dreaming about creating, as you have dreamt, creating intelligent systems? What is the best way to arrive at new breakthrough ideas and carry them through the fire of criticism and past conventional ideas? Ask your questions. Really, your questions are never dumb, and so is the new one way. And don't take no for an answer. If they are really dumb, you will find out quickly by trying an arrow to see that

they are not leading any place. But follow them and try to understand things your way. That is my advice. I don't know if it's going to help anyone. No, that's brilliantly. There is a lot of inertia in science, in academia. It is slowing down science. Yeah, those two words, your way, that's a powerful thing. It's against inertia, potentially, against the flow.

Against your professor. I'll get you a professor. I wrote the book of why in order to democratize common sense. Yes. In order to instill rebellious spirit in students, so they wouldn't wait until the professor Get things right The as you wrote the manifesto of the rebellion

Against the professor in the professor. Yes So looking back at your life of research What ideas do you hope ripple through the next many decades? What do you hope your legacy will be? I already have tombstone Calfs Boy and the fundamental law of Contofactions

That's what it's it's a simple equation. What it can tofectual in terms of a model surgery. That's it, because everything follows from that. If you get that, all the rest, I can die in peace and my student can derive all my knowledge, my mother's mother's will means. The rest follows. Yeah. She did. Thank you so much for talking to me.

I really appreciate it. Well, thank you for being so attentive and instigating. We did it. We did it. The coffee helped. Thanks for listening to this conversation with you, Daya Pearl. And thank you to our presenting sponsor, Cash App. Download it. Use code LEXpodcast.

You'll get $10 and $10 will go to first. A STEM education on profit that inspires hundreds of thousands of young minds to learn and to dream of engineering our future. If you enjoy this podcast, subscribe on YouTube, give it 5 stars on Apple Podcast, support on Patreon, or simply connect with me on Twitter. And now let me leave you with some words of wisdom from Judea Pearl. You cannot answer a question that you cannot ask, and you cannot ask a question that you have no words for.

Thank you for listening and hope to see you next time. Thank you.