See all Lex Fridman transcripts on Youtube

Speaker 1

00:00

The following is a conversation with Jay McClelland, a cognitive scientist at Stanford and 1 of the seminal figures in the history of artificial intelligence and specifically neural networks. Having written the parallel distributed processing book with David Rommelhart, who co-authored the back propagation paper with Jeff Hinton. In their collaborations, they've paved the way for many of the ideas at the center of the neural network based machine learning revolution of the past 15 years. To support this podcast, please check out our sponsors in the description.

Speaker 1

00:36

This is the Lex Friedman Podcast, and here is my conversation with Jay McClelland. You are 1 of the seminal figures in the history of neural networks at the intersection of cognitive psychology and computer science. What to you has, over the decades, emerged as the most beautiful aspect about neural networks, both artificial and biological? The fundamental thing I think about with neural networks is how they allow us to link

Speaker 2

01:08

biology with the mysteries of thought and You know in the when I was first entering the field myself in the late 60s, early 70s, cognitive psychology had just become a field. There was a book published in 67 called Cognitive Psychology. And the author said that the study of the nervous system was only of peripheral interest.

Speaker 2

01:44

It wasn't going to tell us anything about the mind. And I didn't agree with that. I always felt, oh look, I'm a physical being. From dust to dust, ashes to ashes, and somehow I emerged from that.

Speaker 1

02:06

So that's really interesting. So there was a sense with cognitive psychology that in understanding the sort of neuronal structure of things, you're not going to be able to understand the mind. And then your sense is if we study these neural networks, we might be able to get at least very close to understanding the fundamentals of the human mind.

Speaker 2

02:28

Yeah. I used to think, or I used to talk about the idea of awakening from the Cartesian dream. So Descartes thought about these things, right? He was walking in the gardens of Versailles 1 day and he stepped on a stone and a statue moved.

Speaker 2

02:52

And he walked a little further, he stepped on another stone and another statue moved. And he like, why did the statue move when I stepped on the stone? And he went and talked to the gardeners and he found out that they had a hydraulic system that allowed the physical contact with the stone to cause water to flow in various directions, which caused water to flow into the statue and move the statue. And he used this as the beginnings of a theory about how animals act.

Speaker 2

03:28

And he had this notion that these little fibers that people had identified that weren't carrying the blood, were these little hydraulic tubes that if you touch something, there would be pressure and it would send a signal of pressure to the other parts of the system and that would cause action. So he had a mechanistic theory of animal behavior. And he thought that the human had this animal body, but that some divine something else had to have come down and been placed in him to give him the ability to think. Right, so the physical world includes the body in action but it doesn't include thought according to Descartes, right?

Speaker 2

04:19

And so the study of physiology at that time was the study of sensory systems and motor systems and things that you could directly measure when you stimulated neurons and stuff like that. And the study of cognition was something that was tied in with abstract computer algorithms and things like that. But when I was an undergraduate, I learned about the physiological mechanisms. And so when I'm studying cognitive psychology as a first year PhD student, I'm saying, wait a minute, the whole thing is biological, right?

Speaker 1

04:57

You had that intuition right away. That seemed obvious to you.

Speaker 2

05:00

Yeah, yeah.

Speaker 1

05:03

Isn't that magical though that from just the little bit of biology can emerge the full beauty of the human experience? Is it, why is that so obvious to you?

Speaker 2

05:13

Well, obvious and not obvious at the same time. And I think about Darwin in this context too, because Darwin knew very early on that none of the ideas that anybody had ever offered gave him a sense of understanding how evolution could have worked. But he wanted to figure out how it could have worked.

Speaker 2

05:40

That was his goal. And he spent a lot of time working on this idea and coming, you know, reading about things that gave him hints and thinking they were interesting but not knowing why, drawing more and more pictures of different birds that differ slightly from each other and so on, you know, and then then he figured it out. But after he figured it out, he had nightmares about it. He would dream about the complexity of the eye and the arguments that people had given about how ridiculous it was to imagine that that could have ever emerged from some sort of unguided process, that it hadn't been the product of design.

Speaker 2

06:29

And So he didn't publish for a long time, in part because he was scared of his own ideas. He didn't think they could possibly be true. But then by the time the a century rolls around, we all, you know, we understand that evolution, or many people understand or believe that evolution produced, you know, the entire range of animals that there are. And, you know, Descartes' idea starts to seem a little wonky after a while, right?

Speaker 2

07:08

Like, well, wait a minute. There's the apes and the chimpanzees and the bonobos and they're pretty smart in some ways. So what, oh, somebody comes, oh, there's a certain part of the brain that's still different. They don't, there's no hippocampus in the monkey brain.

Speaker 2

07:28

It's only in the human brain. Huxley had to do a surgery in front of many, many people in the late 19th century to show to them there's actually a hippocampus in the chimpanzee's brain. You know? So their continuity of the species is another element that contributes to this idea that we are ourselves a total product of nature.

Speaker 2

08:01

And that to me is the magic and the mystery how nature could actually

Speaker 1

08:15

give rise to organisms that have the capabilities that we have. So it's interesting because even the idea of evolution is hard for me to keep all together in my mind. So because we think of a human time scale, it's hard to imagine, the development of the human eye would give me nightmares too.

Speaker 1

08:36

Because you have to think across many, many, many generations. And it's very tempting to think about kind of a growth of a complicated object. And it's like, how is it possible for that such, such a thing to be built? Because also, me from a robotics engineering perspective, it's very hard to build these systems.

Speaker 1

08:55

How can, through an undirected process, can a complex thing be designed? It seems wrong.

Speaker 2

09:03

Yeah, so that's absolutely right. And a slightly different career path that would have been equally interesting to me would have been to actually study the process of embryological development flowing on into brain development and the exquisite sort of laying down of pathways and so on that occurs in the brain. And I know the slightest bit about that is not my field, but there are fascinating aspects to this process that eventually result in the complexity of various brains.

Speaker 2

09:54

At least 1 thing we're, in the field I think people have felt for a long time, in the study of vision, the continuity between humans and non-human animals has been second nature for a lot longer. I had this conversation with somebody who's a vision scientist and you're saying, oh, we don't have any problem with this. You know, the monkey's visual system and the human visual system, extremely similar up to certain levels, of course, they diverge after a while. But the first, the visual pathway from the eye to the brain and the first few layers of cortex, or cortical areas, I guess 1 would say, are extremely similar.

Speaker 1

10:49

Yeah, so on the cognition side is where the leap seems to happen with humans. That it does seem we're kind of special. And that's a really interesting question when thinking about alien life or if there's other intelligent alien civilizations out there, is how special is this leap?

Speaker 1

11:05

So 1 special thing seems to be the origin of life itself. However you define that, there's a gray area. And the other leap, this is very biased perspective of a human, is the origin of intelligence. And again, from an engineering perspective, it's a difficult question to ask, an important 1, is how difficult does that leap?

Speaker 1

11:28

How special were humans? Did a monolith come down? Did aliens bring down a monolith and some apes had to touch a monolith

Speaker 2

11:37

to get it? It's a lot like Descartes' idea, right?

Speaker 1

11:41

Exactly, but it just seems 1 heck of a leap to get to this level of intelligence.

Speaker 2

11:48

Yeah, and so Chomsky argued that some genetic fluke occurred 100,000 years ago, And just happened that some human, some hominin predecessor of current humans had this 1 genetic tweak that resulted in language. And language then provided this special thing that separates us from all other animals. I think there's a lot of truth to the value and importance of language, but I think it comes along with the evolution of a lot of other related things related to sociality and mutual engagement with others and establishment of, I don't know, rich mechanisms for organizing and understanding of the world, which language then plugs into.

Speaker 1

13:12

Right, so language is a tool that allows you to do this kind of collective intelligence. And whatever is at the core of the thing that allows for this collective intelligence is the main thing. And it's interesting to think about that 1 fluke, 1 mutation could lead to the first crack opening of the door to human intelligence.

Speaker 1

13:37

Like all it takes is 1. Like evolution just kind of opens the door a little bit and then time and selection takes care of the rest.

Speaker 2

13:45

You know, There's so many fascinating aspects to these kinds of things. So we think of evolution as continuous, right? We think, oh, yes, okay, over 500 million years, there could have been this relatively continuous changes.

Speaker 2

14:08

But that's not what anthropologists, evolutionary biologists found from the fossil record. They found hundreds of millions of years of stasis. And then suddenly a change occurs. Well, suddenly on that scale is a million years or something, or even 10 million years.

Speaker 2

14:35

But the concept of punctuated equilibrium was a very important concept in evolutionary biology. And that also feels somehow right about the stages of our mental abilities. We seem to have a certain kind of mindset at a certain age. And then at another age, we like look at that four-year-old and say, oh my God, how could they have thought that way?

Speaker 2

15:07

Piaget was known for this stage theory of child development. You look at it closely and suddenly those stages are so discrete, the transitions, but the difference between the four-year-old and the seven-year-old is profound. And that's another thing that's always interested me is how we, something happens over the course of several years of experience where at some point we reach the point where something like an insight or a transition or a new stage of development occurs. And these kinds of things can be understood in complex systems research.

Speaker 2

15:47

And so evolutionary biology, developmental biology, cognitive development are all things that have been approached in this kind

Speaker 1

15:59

of a way. Yeah, just like you said, I find both fascinating, those early years of human life, but also the early like minutes, days of, from the embryonic development to like how, from embryos you get like the brain, that development, again, from an engineer perspective, it's fascinating. So it's not, so the early, when you deploy the brain to the human world and it gets to explore that world and learn, that's fascinating.

Speaker 1

16:30

But just like the assembly of the mechanism that is capable of learning. That's like amazing. The stuff they're doing with like brain organoids where you can build many brains and study that self-assembly of a mechanism from like the DNA material. That's like, what the heck?

Speaker 1

16:51

You have literally like biological programs that just generate a system, this mushy thing that's able to be robust and learn in a very unpredictable world and learn seemingly arbitrary things or like a very large number of things that enable survival.

Speaker 2

17:14

Yeah, Ultimately, that is a very important part of the whole process of understanding this sort of emergence of mind from brain kind of thing.

Speaker 1

17:27

And the whole thing seems to be pretty continuous. So Let me step back to neural networks for another brief minute. You wrote parallel distributed processing books that explored ideas of neural networks in the 1980s, together with a few folks.

Speaker 1

17:43

But the books you wrote with David Rommelhart, who is the first author on the back propagation paper, which you have hinted. So these are just some figures at the time that were thinking about these big ideas. What are some memorable moments of discovery and beautiful ideas from those early days.

Speaker 2

18:04

I'm gonna start sort of with my own process in the mid-70s and then into the late 70s when I met Jeff Henson and he came to San Diego and we were all together. In my time in graduate schools, I've already described to you, I had this sort of feeling of, okay, I'm really interested in human cognition, but this disembodied sort of way of thinking about it that I'm getting from the current mode of thought about it isn't working fully for me. And when I got my assistant professorship, I went to UCSD and that was in

Speaker 1

18:56

1974.

Speaker 2

18:58

Something amazing had just happened. Dave Rommelhart had written a book together with another man named Don Norman, and the book was called Explorations in Cognition. And it was a series of chapters exploring interesting questions about cognition, but in a completely sort of abstract, non-biological kind of way.

Speaker 2

19:23

And I'm saying, gee, this is amazing. I'm coming to this community where people can get together and feel like they've collectively exploring ideas. And it was a book that had a lot of, I don't know, lightness to it. And Don Norman, who was the more senior figure to Rumelhart at that time, who led that project, always created this spirit of playful exploration of ideas.

Speaker 2

19:55

And so I'm like, wow, this is great. But I was also still trying to get from the neurons to the cognition. And I realized at 1 point, I got this opportunity to go to a conference where I heard a talk by a man named James Anderson, who was an engineer, but by then a professor in a psychology department who had used linear algebra to create neural network models of perception and categorization and memory. And I, just blew me out of the water that 1 could create a model that was simulating neurons, not just engaged in a stepwise algorithmic process that was construed abstractly, but it was simulating remembering and recalling and recognizing the prior occurrence of a stimulus or something like that.

Speaker 2

21:08

For me, this was a bridge between the mind and the brain. I remember I was walking across campus 1 day in 1977, and I almost felt like St. Paul on the road to Damascus. I said to myself, you know, if I think about the mind in terms of a neural network, it will help me answer the questions about the mind that I'm trying to answer.

Speaker 2

21:36

And that really excited me. So I think that a lot of people were becoming excited about that. And 1 of those people was Jim Anderson, who I had mentioned. Another 1 was Steve Grossberg, who had been writing about neural networks since the 60s.

Speaker 2

21:58

And Jeff Hinton was yet another. And his PhD dissertation showed up in an applicant pool to a postdoctoral training program that Dave and Don, the 2 men I mentioned before, Romelhart and Norman, were administering and Romelhart got really excited about Hinton's PhD dissertation. And so Hinton was 1 of the first people who came and joined this group of postdoctoral scholars that was funded by this wonderful grant that they got. Another 1 who is also well-known in neural network circles is Paul Smolensky.

Speaker 2

22:45

He was another 1 of that group. Anyway, Jeff and Jim Anderson organized a conference at UCSD where we were, and it was called Parallel Models of Associative Memory and it brought all the people together who had been thinking about these kinds of ideas in 1979 or 1980. And this began to kind of really resonate with some of Rommelhart's own thinking, some of his reasons for wanting something other than the kinds of computation he'd been doing so far. So let me talk about Rommelhart now for a minute, okay, with that context.

Speaker 1

23:33

Well, let me also just pause, because he said so many interesting things before we go to Rommel-Hart. So first of all, for people who are not familiar, neural networks are at the core of the machine learning, deep learning revolution of today. Jeffrey Hidden, that we mentioned, is 1 of the figures that were important in the history like yourself in the development of these neural networks, artificial neural networks that are then used for the machine learning application.

Speaker 1

23:56

Like I mentioned, the back propagation paper is 1 of the optimization mechanisms by which these networks can learn. And the word parallel is really interesting. So it's almost like synonymous from a computational perspective how you thought at the time about neural networks, that it's parallel computation. Would that be fair to say?

Speaker 2

24:21

Well, yeah, the word parallel in this comes from the idea that each neuron is an independent computational unit, right? It gathers data from other neurons, it integrates it in a certain way, and then it produces a result. And it's a very simple little computational unit, But it's autonomous in the sense that it does its thing, right?

Speaker 2

24:51

It's in a biological medium where it's getting nutrients and various chemicals from that medium. But you can think of it as almost like a little computer in and of itself. The idea is that each, our brains have, oh, look, a hundred or hundreds, almost a billion of these little neurons,

Speaker 1

25:19

right?

Speaker 2

25:21

And they're all capable of doing their work at the same time, so it's like, instead of just a single central processor that's engaged in chug, chug, 1 step after another. We have a billion of these little computational units working at the same time.

Speaker 1

25:42

So at the time, I don't know, maybe you can comment, it seems to me, even still to me, quite a revolutionary way to think about computation relative to the development of theoretical computer science alongside of that, where it's very much like sequential computer. You're analyzing algorithms that are running on a single computer. You're saying, wait a minute, why don't we take a really dumb, very simple computer and just have a lot of them interconnected together and they're all operating in their own little world and they're communicating with each other and thinking of computation that way.

Speaker 1

26:21

And from that kind of computation, trying to understand how things like certain characteristics of the human mind can emerge.

Speaker 2

26:30

Right.

Speaker 1

26:31

That's quite a revolutionary way of thinking, I would say.

Speaker 2

26:35

Well, yes, I agree with you. And there's still this sort of sense

Speaker 1

26:44

of

Speaker 2

26:47

Not sort of knowing how we kind of get all the way there, I think. And this very much remains at the core of the questions that everybody's asking about the capabilities of deep learning and all these kinds of things. But if I could just play this out a little bit, a convolutional neural network or a CNN, which many people may have heard of, is a set of, you could think of it biologically as a set of collections of neurons.

Speaker 2

27:26

Each 1 had, each collection has maybe 10,000 neurons in it but there's many layers, right? Some of these things are hundreds or even a thousand layers deep, but others are closer to the biological brain and maybe they're like 20 layers deep or something like that. So we have, within each layer, we have thousands of neurons or tens of thousands maybe. Well, in the brain, we probably have millions in each layer, but we're getting sort of similar in a certain way, right?

Speaker 2

28:06

And then we think, okay, at the bottom level, there's an array of things that are like the photoreceptors. In the eye, they respond to the amount of light of a certain wavelength at a certain location on the pixel array. So that's like the biological eye. And then there's several further stages going up, layers of these neuron-like units.

Speaker 2

28:30

And you go from that raw input, array of pixels, to the classification, you've actually built a system that could do the same kind of thing that you and I do when we open our eyes and we look around and we see there's a cup, there's a cell phone, there's a water bottle, and these systems are doing that now, right? So they are, in terms of the parallel idea that we were talking about before, They are doing this massively parallel computation in the sense that each of the neurons in each of those layers is thought of as computing its little bit of something about the input simultaneously with all the other ones in the same layer. We get to the point of abstracting that away and thinking, oh, it's just 1 whole vector that's being computed, 1 activation pattern that's computed in a single step, and that abstraction is useful, but it's still that parallel and distributed processing. Each 1 of these guys is just contributing a tiny bit to that whole thing.

Speaker 1

29:45

And that's the excitement that you felt that from these simple things, you can emerge when you add these level of abstractions on it. You can start getting all the beautiful things that we think about as cognition. And so, okay, so you have this conference, I forgot the name already, but it's Parallel and Something Associated with Memory and so on.

Speaker 1

30:05

Very exciting, technical and exciting title. And you started talking about Dave Romahart. So who is this person that was so, You've spoken very highly of him. Can you tell me about him, his ideas, his mind, who he was as a human being, as a scientist?

Speaker 2

30:24

So Dave came from a little tiny town in Western South Dakota. And his mother was the librarian and his father was the editor of the newspaper. And I know 1 of his brothers pretty well.

Speaker 2

30:46

They grew up, there were 4 brothers, and they grew up together, and their father encouraged them to compete with each other a lot. They competed in sports, And they competed in mind games. I don't know, things like Sudoku and chess and various things like that. And Dave was a standout undergraduate.

Speaker 2

31:16

He went at a younger age than most people do to college at the University of South Dakota and majored in mathematics. And I don't know how he got interested in psychology, but he applied to the mathematical psychology program at Stanford and was accepted as a PhD student to study mathematical psychology at Stanford. So mathematical psychology is the use of mathematics to model mental processes.

Speaker 1

31:50

So something that I think these days might be called cognitive modeling, that whole space.

Speaker 2

31:55

Yeah, it's mathematical in the sense that you say if this is true and that is true, then I can derive that this should follow. And so you say, these are my stipulations about the fundamental principles, and this is my prediction about behavior, And it's all done with equations. It's not done with a computer simulation.

Speaker 2

32:19

So you solve the equation, and that tells you what the probability that the subject will be correct on the seventh trial of the experiment is or something like that. So it's a use of mathematics to descriptively characterize aspects of behavior. And Stanford at that time was the place where there were several really, really strong mathematical thinkers who were also connected with 3 or 4 others around the country, who brought a lot of really exciting ideas onto the table. And it was a very, very prestigious part of the field of psychology at that time.

Speaker 2

33:05

So Rumelhart comes into this, he was a very strong student within that program, and he got this job at this brand new university in San Diego in 1967, where he's 1 of the first assistant professors in the Department of Psychology at UCSD. So I got there in 74, 7 years later, and Runghart at that time was still doing mathematical modeling, But he had gotten interested in cognition. He'd gotten interested in understanding. And understanding, I think, remains… what does it mean to understand anyway?

Speaker 2

34:08

It's an interesting sort of curious, like how would we know if we really understood something? But he was interested in building machines that would hear a couple of sentences and have an insight about what was going on. So for example, 1 of his favorite things at that time was, Margie was sitting on the front step when she heard the familiar jingle of the Good Humor Man. She remembered her birthday money and ran into the house.

Speaker 2

34:41

What is Margie doing? Why? Well, there's a couple of ideas you could have, but the most natural 1 is that the good humor man brings ice cream. She likes ice cream.

Speaker 2

34:57

She knows she needs money to buy ice cream, so she's going to run into the house and get her money so she can buy herself an ice cream. It's a huge amount of inference that has to happen to get those things to link up with each other. He was interested in how the hell that could happen. He was trying to build good old-fashioned AI style models of representation of language and content of things like has money.

Speaker 1

35:33

So like formal logic and knowledge bases, like that kind of stuff. So he was integrating that with his thinking about cognition. Yes.

Speaker 1

35:40

The mechanisms cognition, how can they mechanistically be applied to build these knowledge, like to actually build something that looks like a web of knowledge and thereby from there emerges something like understanding.

Speaker 2

35:56

What the

Speaker 1

35:56

heck that is. Yeah, he was

Speaker 2

35:58

grappling, This was something that they grappled with at the end of that book that I was describing, Explorations in Cognition. But he was realizing that the paradigm of good old-fashioned AI wasn't giving him the answers to these questions.

Speaker 1

36:16

By the way, that's called good old fashioned AI now. It wasn't called that

Speaker 2

36:20

at the time. Well, it was. It was beginning to be called that.

Speaker 1

36:23

Oh, because it was from the 60s.

Speaker 2

36:24

Yeah, yeah. By the late 70s, it was kind of old fashioned and it hadn't really panned out, you know? And people were beginning to recognize that.

Speaker 2

36:33

But – and Rumelhart was, you know, like, yeah, he was part of the recognition that this wasn't all working. Anyway, so he started thinking in terms of the idea that we needed systems that allowed us to integrate multiple simultaneous constraints in a way that would be mutually influencing each other. So he wrote a paper that just really, first time I read it, I said, oh, well, you know, yeah, but is this important? But after a while, it just got under my skin.

Speaker 2

37:15

And it was called an interactive model of reading. And in this paper, he laid out the idea that every aspect of our interpretation of what's coming off the page when we read, at every level of analysis you can think of, actually depends on all the other levels of analysis. So what are the actual pixels making up each letter? What do those pixels signify about which letters they are?

Speaker 2

38:00

What do those letters tell us about what words are there? And what do those words tell us about what ideas the author is trying to convey? And so, he had this model where we have these little tiny elements that represent each of the pixels of each of the letters and then other ones that represent the line segments in them, and other ones that represent the letters, and other ones that represent the words. And at that time, his idea was there's this set of experts.

Speaker 2

38:42

There's an expert about how to construct a line out of pixels and another expert about how, which sets of lines go together to make which letters and another 1 about which letters go together to make which words and another 1 about what the meanings of the words are and another 1 about how the meanings fit together, and things like that. And all these experts are looking at this data and they're updating hypotheses at other levels. So the word expert can tell the letter expert, Oh, I think there should be a T there because I think there should be a word the here and the bottom up sort of feature to letter expert could say, I think there should be a T there too. And if they agree, then you see a T, right?

Speaker 2

39:28

And so there's a top down, bottom up interactive process, but it's going on at all layers simultaneously. So everything can filter all the way down from the top as well as all the way up from the bottom. And it's a completely interactive, bidirectional, parallel distributed process.

Speaker 1

39:45

That is somehow because of the abstractions is hierarchical. So like, so there's different layers of responsibilities, different levels of responsibilities. First of all, it's fascinating to think about it in this kind of mechanistic way.

Speaker 1

39:58

So not thinking purely from the structure of a neural network or something like a neural network, but thinking about these little guys that work on letters and then the letters come words and words become sentences. And that's a very interesting hypothesis that from that kind of hierarchical structure can emerge understanding.

Speaker 2

40:21

Yeah, but the thing is, though, I wanna just sort of relate this to earlier part of the conversation. When Romelhart was first thinking about it, there were these experts on the side, 1 for the features and 1 for the letters and 1 for how the letters make the words and so on, and they would each be working, sort of evaluating various propositions about, you know, is this combination of features here going to be 1 that looks like the letter T and so on? And what he realized kind of after reading Hinton's dissertation and hearing about Jim Anderson's linear algebra based neural network models that I was telling you about before was that he could replace those experts with neuron-like processing units which just would have their connection weights that would do this job.

Speaker 2

41:17

So what ended up happening was that Romulhart and I got together and we created a model called the Interactive Activation Model of Letter Perception, which takes these little pixel level inputs, constructs line segment features, letters and words, but now we built it out of a set of neuron like processing units that are just connected to each other with connection weights. So the unit for the word time has a connection to the unit for the letter T in the first position and the letter I in the second position, so on. And because these connections are bidirectional, if you have prior knowledge that it might be the word time, that starts to prime the letters in the features. And if you don't, then it has to start bottom up.

Speaker 2

42:15

But the directionality just depends on where the information comes in first. And if you have context together with features at the same time, they can convergently result in an emergent perception. And that was the piece of work that we did together that sort of got us both completely convinced that this neural network way of thinking was going to be able to actually address the questions that we were interested in as cognitive psychologists.

Speaker 1

42:50

So the algorithmic side, the optimization side, those are all details, like when you first start, the idea that you can get far with this kind of way of thinking, that in itself is a profound idea. So do you like the term connectionism to describe this kind of set of ideas?

Speaker 2

43:07

I think it's useful. It highlights the notion that the knowledge that the system exploits is in the connections between the units, right? There isn't a separate dictionary, there's just the connections between the units.

Speaker 2

43:27

So I already sort of laid that on the table with the connections from the letter units to the unit for the word time. The unit for the word time isn't a unit for the word time for any other reason than it's got the connections to the letters that make up the word time. Those are the units on the input that excite it when it's excited that in a sense represents in the system that there's support for the hypothesis that the word time is present in the input. But it's not, the word time isn't written anywhere inside the model, it's only written there in the picture we drew of the model to say that's the unit for the word time, right?

Speaker 2

44:16

And if somebody wants to tell me, well, how do you spell that word? You have to use the connections from that out to then get those letters, for example.

Speaker 1

44:27

That's such a, that's a counterintuitive idea. We humans want to think in this logic way. This idea of connectionism, it doesn't, it's weird.

Speaker 1

44:41

It's weird that this is how it all works.

Speaker 2

44:43

Yeah, but let's go back to that CNN, right? That CNN with all those layers of neuron-like processing units that we were talking about before, it's going to come out and say, this is a cat, that's a dog. But it has no idea why it said that.

Speaker 2

44:57

It's just got all these connections between all these layers of neurons, like from the very first layer to the, you know, like whatever these layers are, they just get numbered after a while because they somehow further in you go, the more abstract the features are, but it's a graded and continuous process of abstraction anyway. It goes from very local, very specific to much more global, but it's still another pattern of activation over an array of units. And then at the output side, it says it's a cat or it's a dog. And when I open my eyes and say, oh, that's Lex, or, oh, there's my own dog, and I recognize my dog, which is a member of the same species as many other dogs, but I know this 1 because of some slightly unique characteristics.

Speaker 2

45:57

I don't know how to describe what it is that makes me know that I'm looking at Lex or at my particular dog, right? Or even that I'm looking at a particular brand of car. Like I could say a few words about it, but if I wrote you a paragraph about the car, you would have trouble figuring out which car is he talking about, right? So the idea that we have propositional knowledge of what it is that allows us to recognize that this is an actual instance of this particular natural kind has always been something that it never worked, right?

Speaker 2

46:36

You couldn't ever write down a set of propositions for visual recognition. And so in that space, it always seemed very natural that something more implicit, you know, you don't have access to what the details of the computation were in between, you just get the result. So that's the other part of connectionism. You cannot, you don't read the contents of the connections.

Speaker 2

47:03

The connections only cause outputs to occur based on inputs.

Speaker 1

47:09

Yeah, and for us that final layer or some particular layer is very important. The 1 that tells us that it's our dog or like it's a cat or a dog, but you know, each layer is probably equally as important in the grand scheme of things. Like, there's no reason why the cat versus dog is more important than the lower level activations, it doesn't really matter.

Speaker 1

47:33

I mean, all of it is just this beautiful stacking on top of each other. And we humans live in this particular layers. For us, it's useful to survive, to use those cat versus dog, predator versus prey, all those kinds of things. It's fascinating that it's all continuous.

Speaker 1

47:51

But then you then ask, you know, the history of artificial intelligence, you ask, are we able to introspect and convert the very things that allow us to tell the difference to cat and dog into logic, into formal logic. That's been the dream. I would say that's still part of the dream of symbolic AI. And I've recently talked to Doug Lennard, who created Psych.

Speaker 1

48:19

And that's a project that lasted for many decades and still carries a sort of dream in it, right? But we still don't know the answer, right? It seems like a connectionism is really powerful, but it also seems like there's this building of knowledge.

Speaker 2

48:38

And

Speaker 1

48:38

so how do we, how do you square those 2? Like, do you think the connections can contain the depth of human knowledge and the depth of what Dave Romohart was thinking about of understanding?

Speaker 2

48:52

Well, that remains a $64 question. And I-

Speaker 1

48:57

With inflation, that number's higher.

Speaker 2

48:59

Okay, $64,000. Maybe it's the $64 billion question now. I think that from the emergentist side which I place myself on.

Speaker 2

49:23

So I used to sometimes tell people I was a radical eliminative connectionist because I didn't want them to think that I wanted to build like anything into the machine, but I don't like the word eliminative anymore because it makes it seem like it's wrong to think that there is this emergent level of understanding. And I disagree with that. So I think, you know, I would call myself an erratical emergentist connectionist rather than a liminative connectionist, because I want to acknowledge that these higher level kinds of aspects of our cognition are real, but

Speaker 1

50:23

they

Speaker 2

50:26

don't exist as such. And there was an example that Doug Hofstadter used to use that I thought was helpful in this respect. Just the idea that we can think about sand dunes as entities and talk about like how many there are even.

Speaker 2

50:51

But we also know that a sand dune is a very fluid thing. It's a pile of sand that is capable of moving around under the wind and reforming itself in somewhat different ways. And if we think about our thoughts as like sand dunes, as being things that emerge from just the way all the lower level elements sort of work together and are constrained by external forces, then we can say yes, they exist as such, but They also, you know, we shouldn't treat them as completely monolithic entities that we can understand without understanding sort of all of the stuff that allows them to change in the ways that they do. And that's where I think the connectionist feeds into the cognitive.

Speaker 2

51:52

It's like, okay, so if the substrate is parallel distributed connectionist, then it doesn't mean that the contents of thought isn't abstract and symbolic, but it's more fluid maybe than is easier to capture with a set of logical expressions.

Speaker 1

52:15

Yeah, that's a heck of a sort of thing to put at the top of a resume, radical emerginist connectionist. So there is just like you said, a beautiful dance between that, between the machinery of intelligence, like the neural network side of it and the stuff that emerges. I mean, the stuff that emerges seems to be, I don't know what that is.

Speaker 1

52:44

It seems like maybe all of reality is emergent. What I think about, this is made most distinctly rich to me when I look at cellular automata, look at game of life, that from very, very simple things, very rich, complex things emerge that start looking very quickly like organisms that you forget how the actual thing operates. They start looking like they're moving around, they're eating each other. Some of them are generating offspring.

Speaker 1

53:20

You forget very quickly. And it seems like maybe it's something about the human mind that wants to operate in some layer of the emergent and forget about the mechanism of how that emergence happens. So just like you are in your radicalness, also it seems like unfair to eliminate the magic of that emergence. Like eliminate the fact that that emergence is real.

Speaker 2

53:48

Yeah, no, I agree. That's why I got rid of eliminative, right?

Speaker 1

53:52

Eliminative, yeah.

Speaker 2

53:53

Yeah, because it seemed like that was trying to say that it's all completely like...

Speaker 1

54:01

An illusion of some kind.

Speaker 2

54:03

Well, who knows whether there aren't some illusory characteristics there. And I think that philosophically, many people have confronted that possibility over time. But it's still important to accept it as magic, right?