Lex Fridman Podcast - Yann LeCun: Deep Learning, Convolutional Neural Networks, and Self-Supervised Learning
Episode Date: August 31, 2019Yann LeCun is one of the fathers of deep learning, the recent revolution in AI that has captivated the world with the possibility of what machines can learn from data. He is a professor at New York Un...iversity, a Vice President & Chief AI Scientist at Facebook, co-recipient of the Turing Award for his work on deep learning. He is probably best known as the founder of convolutional neural networks, in particular their early application to optical character recognition. 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 iTunes or support it on Patreon.
Transcript
Discussion (0)
The following is a conversation with Yalekun.
He's considered to be one of the fathers of deep learning,
which, if you've been hiding under Iraq,
is the recent revolution in AI that's captivated the world
with the possibility of what machines can learn from data.
He's a professor in New York University of ICE President
and Chief AI scientist at Facebook
and co-recipient the Turing Award for his work on deep learning.
He's probably best known as the founding father of convolutional neural networks.
In particular, their application to optical character recognition and the famed M-NIST
data set.
He is also an outspoken personality, unafraid to speak his mind in a distinctive French accent
and explore provocative ideas
both in the rigorous medium of academic research and the somewhat less rigorous medium of Twitter
and Facebook. This is the Artificial Intelligence Podcast. If you enjoy it, subscribe on YouTube,
give it 5 stars and iTunes, support it on Patreon, or simply to click me on Twitter at Lex Friedman's about the F-R-I-D-M-A-N.
And now here's my conversation with Jan LeCoon. You said that 2001 Space Odyssey is one of your favorite movies.
Hal 9000 decides to get rid of the astronauts for people who haven't seen the movie spoiler
alert because he believes that the they will interfere with the mission.
Do you see how is flawed in some fundamental way, or even evil, or did he do the right thing?
Neither. There's no notion of evil in that context, other than the fact that people die.
But it was an example of what people call
value misalignment, right?
You give an objective to a machine and
the machine tries to achieve this objective.
If you don't put any constraints on
this objective, don't kill people and don't do things like this.
The machine given the power will do stupid things just to
achieve this objective or damaging just to achieve this objective or damaging
things to achieve this objective.
It's a little bit like, we are used to this in a context of human society.
We put in place laws to prevent people from doing bad things because spontaneously they
would do those bad things.
So we have to shape their cost function, their objective
function, if you want to laws to kind of correct and education obviously, to sort of correct
for those. So maybe just pushing a little further on that point, how there's a mission,
there's a fuzziness around the ambiguity around what the actual mission is, but do you think
that there will be a time from a utilitarian perspective
where an AI system, where it is not misalignment,
where it is alignment for the greater good of society,
that an AI system will make decisions that are difficult?
Well, that's the trick.
I mean, eventually it will have to figure out how to do this.
And again, we're not starting from scratch
because we've been doing this with humans for millennia.
So designing objective functions for people
is something that we know how to do.
And we don't do it by programming things,
although the legal code is called code.
That tells you something.
It's actually the design of an objective function.
That's really what legal code is.
It tells you, here is what you can do,
here is what you can't do.
If you do it, you pay that much.
That's an objective function.
There is this idea somehow that it's
a new thing for people to try to design
objective functions that are aligned with the common good.
But no, we've been writing laws from millennia and that's exactly what it is.
So that's where the science of law-making and computer science will come together.
We'll come together.
So there's nothing special about how our AI systems is just the continuation of tools used
to make some of these difficult ethical judgments
that laws make.
Yeah, and we have systems like this already
that make many decisions for ourselves in society
that need to be designed in a way that they
like rules about things that sometimes have bad side effects.
And we have to be flexible enough
about those rules so that they can be broken when it's obvious that they shouldn't be applied.
So you don't see this on the camera here, but all the decoration in this room is all pictures from
2001 and space audices. Wow. Then by accident or is there a lot? It's not by accident. It's by design.
or is there a lot? It's not by accident, it's by design.
Oh wow.
So if you were to build how 10,000,
so an improvement of how 9,000, what would you improve?
Well, first of all, I wouldn't ask you
to hold secrets and tell lies
because that's really what breaks it in the end.
That's the fact that it's asking,
it's health questions about the purpose of the mission. And it's, you know,'s the fact that it's asking, it's health questions about the purpose of
the mission. And it's, you know, pieces things together that it's heard, you know, all the secrecy
of the preparation of the mission and the fact that it was discovery on the lunar surface that
really was kept secret. And one part of how's memory knows this and the other part is,
does not know it and is supposed to not tell anyone and that creates an internal conflict.
Do you think there's never should be a set of things that an AI system should not be allowed?
Like a set of facts that should not be shared with the human operators?
Well, I think no, I think that I think it should be a bit like in the design of
No, I think it should be a bit like in the design of autonomous AI systems, there should be the equivalent of the oath that he for credit.
That's a credit also.
That doctor is a sign of too.
So there's certain things, certain rules that you have to abide by.
And we can sort of hardwire this into our machines
to kind of make sure they don't go.
So I'm not, you know, an advocate of the $3.00
over robotics, you know, the azimuth kind of thing,
because I don't think it's practical,
but, you know, some level of limits.
But to be clear, this is not,
these are not questions that are kind of re-worth asking today because we just don't have the technology to do this.
We don't have autonomous internal machines. We have intelligent machines, so am I intelligent machines that are very specialized, but they don't really sort of satisfy an objective. They're just kind of trained to do one thing. So until we have some idea for
design of a full-fledged autonomous intelligent system, asking the question of how we design
this objective, I think it's a little too abstract.
It's a little too abstract. There's useful elements to it in that it helps us understand
our own ethical codes, humans.
So even just as a thought experiment,
if you imagine that an AGI system is here today,
how would we program it?
Is it a kind of nice thought experiment of constructing?
How should we have a law,
have a system of laws for us humans?
It's just a nice practical tool.
And I think there's echoes of that idea
too in the AI systems we have today that don't have to be that intelligent. Yeah, like autonomous
vehicles. These things start creeping in that they're worth thinking about, but certainly they
shouldn't be framed as how. Yeah. Looking back, what is the most, I'm sorry if it's a silly question, but what is the most
beautiful or surprising idea in deep learning or AI in general that you've ever come across?
So personally, when you said back and just had this kind of, that's a pretty cool moment.
That's nice.
Well, surprising.
I don't know if it's an idea, rather than a sort of empirical fact, the fact that you can
build gigantic neural nets, try them on relatively small amounts of data relatively with
to castigrate and descent, and that actually works.
Breaks everything you read in every textbook,
every pre-depriving textbook that I told you,
you need to have fewer parameters
than you have data samples.
If you have a non-convex objective function,
you have no guarantee of convergence,
all those things that you read in textbook
and they tell you stay away from this.
And they're all wrong.
As each number of parameters, non-convex, and somehow, which is very relative to the number
of parameters data, it's able to learn anything.
Right.
Does that still surprise you today?
Well, it was kind of obvious to me before I knew anything that this is a good idea, and
then it became surprising that it worked because I started reading those textbooks.
Okay.
So you talked to the intuition of why I was obviously
if you remember.
Well, okay, so the intuition was,
it's sort of like those people in the late 19th century
who proved that heavier than air flight was impossible.
And of course you have birds. They do fly.
On the face of it,
it's obviously wrong as an empirical question.
We have the same thing that we know that the brain works.
We don't know how, but we know it works.
We know it's a large network of neurons in interaction,
and that learning takes place by changing the connection.
Getting this level of inspiration
without copying the details,
but sort of trying to derive basic principles,
you know, that kind of gives you a clue
as to which direction to go.
There's also the idea somehow that I've been convinced of
since I was on the undergrad that, even before,
that intelligence is inseparable from learning.
So the idea somehow that you can create
an intelligent machine by basically programming,
for me was an on-starter from the start.
Every intelligent entity that we know about arrives at
this intelligence through learning.
So learning, machine learning was completely obvious path.
Also because I'm lazy, so you know, kind of,
is automate basically everything and learning
is the automation of intelligence.
So do you think, so what is learning
then what what falls under learning?
Because do you think of reasoning as learning?
Well, reasoning is certainly a consequence of learning as well, just like other functions
of the brain.
The big question about reasoning is how do you make reasoning compatible with gradient-based
learning?
Do you think neural networks can be made to reason?
Yes, there is no question about that.
Again, we have a good example, right?
The question is, how?
So the question is how much prior structure you have to put into the neural net so that
something like human reasoning will emerge from it, you know, from learning.
Another question is, all of our kind of model of what reasoning is that are based on logic are discrete and
are therefore incompatible with great and based learning. And I'm a very strong believer in this idea of great and based learning. I don't believe that
all the types of learning that don't use kind of gradient information if you want. So you don't like discrete mathematics
you don't like anything discrete. Well, that's it's not that I don't like it, it's just that it's incompatible with learning
and I'm a big fan of learning, right?
So in fact that's perhaps one reason why deep learning has been kind of looked at with
suspicion by a lot of computer scientists because the math is very different.
The math that you use for deep learning, you know, what we're kind of, as more to do with
you know, sabonetics, the kind of math you do in electric
engineering, then the kind of math you do in computer science.
And, you know, nothing in machine learning is exact, right?
Computer science is all about sort of, you know, obviously, compulsive attention to details
of like, you know, every index has to be right.
And you can prove that algorithm is correct, right?
Machine learning is
the science of solopiness really. That's beautiful. So, okay, maybe let's feel around in the dark
of what is a neural network that reasons or a system that is works with continuous functions
that is works with continuous functions
that's able to do build knowledge. However we think about reasoning,
build on previous knowledge, build on extra knowledge,
create new knowledge, generalize outside of any training set
ever built, what does that look like?
If, yeah, maybe they have inklings of thoughts
of what that might look like.
Yeah, I mean, yes or no.
If I had precise ideas about this,
I think, you know, we'll be building it right now.
But, and there are people working on this,
who's main research interest is actually exactly that, right?
So, what you need to have is a working memory.
So, you need to have some device, memory. So you need to have some device if you want,
some subsystem that can store a relatively large number
of factual episodic information for a reasonable amount of time.
So in the brain, for example,
there are three main types of memory.
One is the sort of memory of the state of your cortex.
And that sort of disappears within 20 seconds.
You can't remember things for more than about 20 seconds or a minute,
if you don't have any other form of memory.
The second type of memory, which is longer term, the short term,
is the hippocampus.
So you can, you know, you came into this building, you remember where the, where the exit is,
where the elevators are.
You have some map of that building that's stored in your hippocampus.
You might remember something about what I said, you know, a few minutes ago.
I forgot it all already, for a while.
It's been erased, but, you know, but that, that would be in your be in your hippocampus. And then the longer term memory is in the synapses.
So what you need if you want to system that's
keep people reasoning is that you want the hippocampus
like thing.
And that's what people have tried to do with memory networks
and neural-turing machines and stuff like that.
And now with transformers, which have sort of a memory in there,
kind of self-attention system, you can think of it this way.
So, so that's one element you need.
Another thing you need is some sort of network that can access this memory,
get an information back, and then kind of crunch on it it and then do this iteratively multiple times
because a chain of reasoning is a process by which you update your knowledge about the
set of the world about what's going to happen, etc.
And that has to be this recurrent operation basically.
And you think that kind of, if you think about a transformer, so that seems to be this sort of recurrent operation, basically. And you think that kind of, if you think about a transformer,
so that seems to be too small to contain the knowledge
that's to represent the knowledge
that's contained Wikipedia, for example.
Well, a transformer doesn't have this idea of recurrence.
It's got a fixed number of layers,
and that's a number of steps that will limit
basically its representation.
But recurrence would build on the knowledge somehow.
I mean, it would evolve the knowledge
and expand the amount of information,
perhaps, or useful information within that knowledge.
But is this something they just can emerge with size?
Because it seems like everything we have now is just...
No, it's not clear.
I mean, how are you access and write into an associated memory inefficient way?
I mean, so the original memory network maybe had something like the right
architecture, but if you try to scale up a memory network so that the memory
contains all Wikipedia, it doesn't quite work.
Right.
So there's a need for new ideas there.
But it's not the only form of reasoning.
So there's another form of reasoning, which
is very classical also in some types of AI.
And it's based on, let's call it energy minimization.
So you have some sort of objective,
some energy function that represents the quality or the negative quality.
Energy goes up when things get bad and they get low when things get good.
So let's say you want to figure out what gestures do I need to do to grab an object or work out the door. If you have a good model of your own body,
a good model of the environment,
using this kind of energy minimization,
you can make, you can do planning.
And it's in optimal control,
it's called model predictive control.
You have a model of what's gonna happen
in the world as consequence of your actions.
And that allows you to buy energy minimization,
figure out a sequence of action that
optimizes a particular objective function, which
measures minimizes the number of times
you're going to hit something, and the energy
you're going to spend doing the gesture and et cetera.
So that's a formal reasoning.
Planning is a formal reasoning.
And perhaps what led to the ability of humans to reason is the fact that
or species that appear before us had to do some sort of planning to be able to hunt and survive
and survive the winter in particular. And so it's the same capacity that you need to have.
So in your intuition is if we look at expert systems and encoding knowledge as logic
systems, as graphs, in this kind of way, is not a useful way to think about knowledge.
Graphs are little brittle or logic representation.
So basically, you know, variables that have values and then constrained between them that are represented by rules,
is little too rigid and too brittle, right? So one of the, you know, some of the early efforts in that respect,
were to put probabilities on them. So a rule, you know, if you have this in that same term, you know, you have this disease with that probability and you should prescribe that antibiotic
with that probability.
That's the mycine system from the 70s.
And that's what that branch of AI led to,
based on networks and graphical models and causal inference
and variational method.
So there is, I mean, certainly a lot of interesting work
going on in this area. The main issue with this is knowledge acquisition. How do you reduce
a bunch of data to a graph of this type?
Yeah, we're lives in the expert on the human being to encode, to add knowledge. And that's
essentially in practical. Yeah.
So that's a big question.
The second question is, do you want
to represent knowledge as symbols?
And do you want to manipulate them with logic?
And again, that's incompatible with learning.
So one suggestion with Jeff Hinton
has been advocating for many decades
is replace symbols by
Vactors think of it as pattern of activities in a bunch of
neurons or units or whatever you want to call them and replace logic by
Continuous functions Okay, and that becomes now compatible. This is a very good set of ideas
By written in a paper about 10 years ago by Leon Batou on who is
here at Facebook. The title of the paper is for machine learning to machine reasoning
and his idea is that a learning system should be able to manipulate objects that are in
a space and then put the result back in the same space. So it's this idea of working memory, basically.
And it's very enlightening.
And in a sense, that might learn something
like the simple expert systems.
I mean, you can learn basic logic operations there.
Yeah, quite possibly.
Yeah.
This big debate on how much prior structure you
have to put in for this kind of stuff to emerge. That's the debate I have with Gary Marcus and people like that. Yeah. This big debate on sort of how much prior structure you have to put in for this kind of stuff to emerge.
That's the debate I have with Gary Marcus and people like that.
Yeah. Yeah. So, and the other person, so I just talked to Judea Pearl,
from the mention causal inference world. So his worry is that the current neural networks are not able to
current neural networks are not able to learn what causes what causal inference between things.
So I think is right and wrong about this.
If he's talking about the sort of classic type of neural nets, people didn't worry too much
about this.
But there's a lot of people now working on causal inference, and there's a paper that
just came out last week by Leon Boutouou among others, the videopass, by Zon
Pacheva, the people. Exactly on that problem of how do you
kind of, you know, get an neural net to sort of pay attention
to real causal relationships, which may also solve issues
of bias in data, can things like this. So I'd like to read that paper because that ultimately the challenge there's also seems
to fall back on the human expert to ultimately decide causality between things.
People are not very good at establishing causality, first of all.
So first of all, you talk to a physicist and physicists actually don't believe in causality, first of all. So first of all, you talk to physicists, and physicists actually don't believe in causality
because look at all the basic laws of macro physics
are time reversible.
So there's no causality.
There are times not real.
It's as soon as you start looking at
macroscopic systems where there is unpredictable randomness
where there is clearly an hour of time,
but it's a big mystery in physics actually,
well, how that emerges.
Is it emergent, or is it part of the,
on the metal fabric of reality, yeah?
Or is it bias of intelligent systems
that, you know, because of the second law of thermodynamics,
we perceive a particular hour of time,
but in fact, it's kind of arbitrary, right?
So yeah, physicists, mathematicians, they don't care about care about I mean the math doesn't care about the flow of time
Well, certainly certainly macrophysics doesn't
People themselves are not very good at establishing
causal
causal relationships if you ask is I think it was in one of similar papers book on on
Like children learning you know, he studied with Jean-Pierre Géin,
he's the guy who co-authored the book Perception
with Marvin Minsky that kind of killed
the first wave of neural nets.
But he was actually a learning person.
He, in the sense of studying learning in humans
and machines, that's why he got interested in Perception.
And he wrote that, if you ask a little kid about what
is the cause of the wind, a lot of kids will say,
they will think for a while, and they'll say, oh,
the branches and the trees, they move,
and that creates wind.
So they get the causal relationship backwards.
And it's because they are understanding
of the world and intuitive physics.
It's not that great.
I mean, these are like four or five year old kids.
It gets better and then you understand that this can be, right?
But there are many things which we can, because of our common sense,
understanding of things, what people call common sense.
Yeah.
And we're understanding of physics.
We can, there's a lot of stuff that we can figure out causality, even with diseases.
We can figure out what's not causing what, often.
There's a lot of mystery, of course, but the idea is that you should be able to encode
that into systems because it seems unlikely to be able to figure that out themselves.
Well, whenever we can do intervention, but all of humanity has been completely diluted for millennia, probably since existence,
about a very, very wrong causal relationship
where whatever you can explain,
you're attributed to some deity, some divinity, right?
And that's a cup out.
That's a way of saying, like, I don't know the cause,
so God did it, right?
So you mentioned Marvin Minsky and the irony of, you know, maybe causing the first
day I went to you were there in the 90s, you were there in 80s, of course. In the 90s,
what do you think people lost faith in deep learning? In the 90s and founded again, a decade
later, over a decade later. Yeah, it wasn't called deep learning yet, it was just called neural nets.
Yeah, they lost interest.
I mean, I think I would put that around 1995, at least the machine learning community.
There was always a neural net community, but it became kind of disconnected from sort of mainstream
machine learning if you want.
There were, it was basically actually a engineering that kept at it and computer science gave up on neural nets.
I don't know.
I was too close to it to really
sort of analyze it with sort of a
unbiased eye if you want, But I would make a few guesses.
So the first one is, at the time, neural nets
were, it was very hard to make them work.
In a sense that you would implement backprop
in your favorite language.
And that favorite language was not Python.
It was not MATLAB, it was not
any of those things because they didn't exist, right? You had to write it in Fortranos
C or something like this, right? So you would experiment with it, you would probably make
some very basic mistakes, like you know, barely initialize your weights, make the network too
small because you're ready in the textbook, you know, you don't want to any parameters,
right? And of course, you know, and you would train on XOR because you're ready in the textbook. You don't want too many parameters.
Of course, you would train on XOR because you didn't have any other dataset to trade on.
Of course, it works after time.
So you would say, give up.
Also, you would train it with batch gradient, which isn't sufficient.
There was a lot of bad metrics that you had to know to make those things work, or you
had to reinvent.
A lot of people just didn't
and they just couldn't make it work.
So that's one thing.
The investment in software platform
to be able to kind of, you know,
display things, figure out why things don't work,
kind of get a good intuition for how to get them to work,
have inner flexibility so you can create, you know,
network architecture is a lot of convolutional nets
and stuff like that.
It was hard. I mean, you had to write everything from scratch.
And again, you didn't have any Python or MATLAB or anything, right?
So I read that, sorry to interrupt,
but I read that you wrote in LISP,
the first versions of Lynette with the convolutional networks,
which by the way, one of my favorite languages.
That's how I knew your legit, touring award, whatever, this is what the program
done list, that's still my favorite language.
But it's not that we programmed in Lisp,
it's that we had to write our Lisp interpreter.
Okay, because it's not like we used one that existed.
So we wrote our Lisp interpreter that we hooked up
to backhand library that we wrote also
for sort of neural-nade computation.
And then after a few years around 1991,
we invented this idea of basically having modules
that know how to forward propagate and back propagate gradients
and then interconnecting those modules in a graph.
Lombard 2 had made proposals on this about this in the late 80s
and we were able to implement this using a list system.
Eventually, we wanted to use that system to make on this about this in the late 80s and we're able to implement this using our list system.
Eventually, we wanted to use that system to make
build production code for character recognition at Bell Labs. So we actually wrote a compiler for that list interpreter so that Patricia Mard, who is now Microsoft, did the bulk of it with
Leon and me. And so we could write our system in this and then compile to C and then we'll have
a self-contained
complete system that could kind of do the entire thing.
Neither PyTorch nor TensorFlow can do this today.
Yeah, OK.
It's coming.
Yeah.
I mean, there's something like that in PyTorch
called Torch Script.
And so we had to write to a list of the triplers,
we had to write to a list compiler, we had to write all this meterplier, we had to write all this compiler,
we had to invest a huge amount of effort to do this.
And not everybody, if you don't completely believe
in the concept, you're not going to invest
the time to do this.
Now, at the time also, what today,
this would turn into Torch or PyTorch or TensorFlow,
or whatever, we'd put it in open source, everybody would use it
and realize it's good.
Back before 1995, working at AT&T,
there's no way the lawyers would let you release anything
in open source of this nature.
And so we could not distribute our code really.
And at that point, and started going
to million tangents, but on that point,
I also read that there was some almost pat, like a patent
on commercial and you know, that work.
Yes.
But it was.
So that, first of all, I mean, just who actually, that ran out, that, thankfully, 2007,
2007.
What, can we, can we just talk about that for, I know you're a Facebook, but you're
also in NYU, and what does it mean to patent ideas like these software ideas, essentially,
or what are mathematical ideas, or what are they?
Okay.
So, they're not mathematical ideas, so there are algorithms. And there was a period
where the US patent office would allow the patent of software as long as it was embodied.
The Europeans are very different. They don't quite accept that. They have a different concept, but
I don't, I don't, I mean, I never actually strongly believed in this, but I don't believe in
this kind of patent. Facebook basically doesn't believe in this kind of patent.
Facebook basically doesn't believe in this kind of patent.
Google
five patents because they've been burned with Apple.
And so now they do this for defensive purpose, but usually they say, we're not going to see you if you're in French.
Facebook has a similar policy, they say,
we've had patterns on certain things.
For defensive purpose, we're not going to see you
if you're in French, when I see you through us.
So the industry does not believe in patterns.
They are there because of the legal landscape
and various things, but I don't really believe
in patterns for this kind of stuff.
So that's a great thing.
So I'll tell you a more story actually.
So what happens was the first pattern
about convolutional net was about the early version
of convolutional net that didn't have separate
pulling layers.
It had convolutional layers with tried more than one
if you want.
And then there was a second one on convolutional nets with separate pulling layers,
turning it back up.
And there were files in 89 and 90 years,
something like this.
At the time, the life of a pattern was 17 years.
So here's what happened over the next few years,
is that we started developing character recognition
technology around convolutional nets.
And in 1994, a check reading system was deployed
in ATM machines.
In 1995, it was for large check reading machines
in back offices, et cetera.
And those systems were developed by an engineering group
that we were collaborating with at AT&T,
and they were commercialized by NCR, which at the time was a subsidiary of AT&T.
Now, AT&T is played up in 1996, early 1996.
The lawyers just looked at all the patents and distributed the patents among the various companies.
They gave the commercial net patent to NCR because they were actually selling products that used it.
But nobody at NCR had any idea what the commercial net was.
Yeah.
Okay.
So between 1996 and 2007,
there's a whole period until 2002 where I didn't actually work on machine learning or
commercial net. I resumed working on this around 2002.
And between 2002 and 2007, I was working on them
crossing my finger that nobody at NCR would notice
and nobody noticed.
Yeah, and I hope that this kind of somewhat,
as you said, lawyers aside relative openness
of the community now will continue.
It accelerates the entire progress of the industry.
And the problems that Facebook and Google and others
are facing today is not whether Facebook or Google
or Microsoft or IBM or whoever is ahead of the other.
It's that we don't have the technology to build these things we want to build.
We want to build intelligent virtual assistants that have common sense.
We don't have monopoly on good ideas for this.
We don't believe we do. Maybe others believe they do, but we don't. If a startup tells you they have the secret to
human level intelligence and common sense, don't believe them. They don't. And it's going to take
the entire work of the world research community for a while to get to the point where you can go
often in each of those companies can start to build things on this. We're not there yet. of the world research community for a while, to get to the point where you can go often
in each of those companies
can start to build things on this.
We're not there yet.
It's absolutely, and this calls to the gap
between the space of ideas
and the rigorous testing of those ideas
of practical application that you often speak to.
You've written advice saying,
don't get fooled by people who claim to have a solution
to artificial general intelligence who claim to have an AI system that works just like the
human brain or who claim to have figured out how the brain works. Ask them what the error
rate they get on M-nist or ImageNet. So this is a little dated, by the way. I mean, five years. Yes. Who's counting.
OK.
But I think your opinion is to amnest an image that, yes,
maybe dated, there may be new benchmarks, right?
But I think that philosophy is when you still
and somewhat hold that benchmarks and the practical testing,
the practical application is where
you really get to test the ideas.
Well, it may not be completely practical.
Like, for example, you know, it could be a toy dataset, but it has to be some sort of
task that the community as a whole has accepted as some sort of standard, you know, kind of
benchmark if you want.
It doesn't need to be real.
So, for example, many years ago here at Fair, people, you know, just in Western Antoine
Borne, and a few others proposed the Babbitas tasks, which were kind of a toy problem to test the ability of
machines to reason actually to access working memory and things like this.
And it was very useful, even though it wasn't a real task.
M-list is kind of halfway a real task.
So you know, toy problems can be very useful.
It's just that I was really struck by the fact that a lot of people,
particularly a lot of people, we need to invest.
We'll be fooled by people telling them,
oh, we have, you know, the algorithm of the cortex and you should give us 50 million.
Yes, absolutely.
So there's a lot of people who try to take advantage of the hype for business reasons and so on.
But let me sort of talk to this idea that new ideas,
the ideas that push the field forward may not yet have a benchmark.
Or it may be very difficult to establish a benchmark.
I agree. That's part of the process.
Establishing benchmarks is part of the process.
So what are your thoughts about?
So we have these benchmarks on around stuff
we can do with images from classification to captioning
to just every kind of information
you can pull off from images in the surface level.
There's audio, data sets, there's some video.
What can we start, natural language?
What kind of stuff, what kind of benchmarks do you see?
They start creeping on to more something like intelligence,
like reasoning, like maybe you don't like the term,
but AGI, echoes of that kind of organization.
A lot of people are working on interactive environments
in which you can train and test intelligence systems. So there, for example,
you know, it's the classical paradigm of super-visioning is that you
you have a data set, you partition it into a training set,
validation set, test set, and there's a clear protocol, right?
But why if the that assumes that the samples are
statistically independent,
you can exchange them,
the order in which you see them,
doesn't matter, you know, things like that.
But what if the answer you give determines the next sample you see,
which is the case, for example, in robotics, right?
You robot does something and then it gets exposed to a new room,
and depending on where it goes,
the room will be different.
So that's the, that creates the exploration problem.
The, what if the samples, so that creates also a dependency between samples, right?
To, you, if you move, if you can only move in, in space,
the next sample you're going to see is going to be probably in the same building,
most likely.
So, so the, so the, all the assumptions about the validity of this training set
set are put as its break. Whenever a machine can take an action that has an
influence in the world and it's what is going to see. So people are sending up
artificial environments where that takes place, right? The robot runs around a 3D
model of a house and can interact with objects and things like this.
So you do robotics by simulation, you have those, you know, opening a gym type thing or
Mujoko kind of simulated robots and you have games, you know, things like that.
So that's where the field is going really, this kind of environment.
Now back to the question of a GI, like, I don't like
the term a GI, because it implies that human intelligence is general, and human intelligence
is nothing like general. It's very, very specialized. We think it's general, we like to think
ourselves as having general intelligence, we don't, we're very specialized. We're only slightly
more general than... Why does it feel general?
So you kind of, the term general,
I think what's impressive about humans
is ability to learn, as we were talking about learning,
to learn in just so many different domains.
It's perhaps not arbitrarily general,
but just you can learn in many domains
and integrate that knowledge somehow.
Okay.
The knowledge persists.
So let me take a very specific example.
Yes.
It's not an example.
It's more like a quasi-methodical demonstration.
So you have about one million fibers coming out of one of your eyes, okay, two million total.
But let's let's talk about just one of them.
It's one million nerve fibers, your optical nerve.
Let's imagine that they are binary, so they can be active or inactive.
So the input to your visual cortex is 1 million bits.
Now they're connected to your brain in a particular way, and your brain has connections that
are kind of a little bit like a convolutional network, they're kind of local, you know, in
space and things like this.
Now imagine I play a trick on you.
It's a pretty nasty trick I admit.
I cut your optical nerve and I put a device
that makes a random perturbation of a permutation
of all the nerve fibers.
So now what comes to your brain is a fixed
but random permutation of all the pixels.
There's no way in hell that your visual cortex, even if I do this to you in infancy,
will actually learn vision to the same level of quality that you can.
Got it. And you're saying there's no way you've relearned that?
No, because now two pixels that are nearby in the world well end up in very different places in your visual cortex.
And your neurons there have no connections with each other
because they only connect it locally.
So this whole, our entire, the hardware is built
in many ways to support the locality of the real world.
Yeah.
Yes, that's specialization.
Yeah, but it's still pretty damn impressive.
So it's not perfect generalization.
It's not even close.
No, no, it's not even close. It's not at all. Yes, not. It's still pretty damn impressive. So it's not perfect generalization. It's not even close. No, it's not even close.
It's not at all.
Yes, not.
It's specialized.
So how many Boolean functions?
So let's imagine you want to train your visual system
to recognize particular patterns of those 1 million bits.
So that's a Boolean function.
Either the pattern is here or not here.
There's a 2A classification with 1 million binary inputs.
How many such Boolean functions are there?
Okay.
You have two to the 1 million combinations of inputs.
For each of those, you have an output bit.
And so you have two to the 1 million Boolean functions of this type, okay,
which is an unimaginably large number. How many of those functions can actually be computed
by your usual cortex? And the answer is a tiny tiny tiny tiny tiny tiny tiny sliver,
like an enormously tiny sliver. So we are ridiculously specialized. Okay, but that's an argument against the word general.
I think there's a, I agree with your intuition, but I'm not sure it seems the brain is impressively
It's impressively capable of adjusting to things. So it's because we can't imagine tasks that are outside of our comprehension.
So we think we're a general, because we're a general of all the things that we cannot
comprehend.
But there is a huge world out there of things that we have no idea.
We call that heat, by the way.
Heat. So at this physicist call that heat, or they call it entropy, which is...
That's true.
You know, you have a...
...thing full of gas, right?
...close system for gas, right?
Close on or close. It has, you know, pressure, it has...
...temperature, it has temperature, and you can write equations, pv equal and
on t, things like that. When you reduce the volume, the temperature goes up,
the pressure goes up, things like that, for perfect gas at least. Those are the
things you can know about that system. And it's a tiny, tiny number of bits
compared to the complete information of the state of about that system. And it's a tiny, tiny number of bits compared to the complete
information of the state of the entire system, because the state of the entire system will give you
the position and momentum of every, every molecule of the gas. And what you don't know about it is
the entropy and you interpret it as heat. The energy contained in that thing is what we call heat. Now, it's very possible
that in fact there is some very strong structure in how those molecules are moving, it's just
that they are in a way that we are just not wired to perceive.
Yeah, we're ignorant to it. And there's in your infinite amount of things we're not
wired to perceive. Yeah.
And you're right, that's a nice way to put it. We're general to all the things we can imagine,
which is a very tiny subset of all things that are possible.
It's like CommaGolf complexity or the CommaGolf
is chitine someone of complexity.
You know, every bit string or every integer is random,
except for all the ones that you can actually write down.
Yeah, okay, so beautiful, but you know, so we can just call it artificial intelligence. We don't need to have a general stamina level. Human level intelligence is a good one.
You know, you'll start anytime you touch human, it gets interesting because,
Any time you touch human, it gets interesting because, you know, it's because we attach ourselves to human and it's difficult to define what human intelligence is.
Nevertheless, my definition is maybe a damn impressive intelligence.
Okay, damn impressive demonstration of intelligence, whatever. And so on that topic, most successes in deep learning have been in supervised learning.
What is your view on supervised learning?
Is there hope to reduce involvement of human input and still have successful systems
that have practically used.
Yeah, I mean, there's definitely a hope.
It's more than a hope actually.
It's mounting evidence for it.
And that's basically all I do,
like the only thing I'm interested in at the moment is,
I call it self-supervised running, not unsupervised.
Because unsupervised running is a loaded term.
People who know something about machine running, tell you, so you're doing clustering or PCA, which is not loaded term. People who know something about machine learning, you know, tell you,
so you're doing clustering or PCA, right? She's not the case. And the white public, you know,
when you say unsupervised learning, oh my god, you know, machines are going to learn about themselves
and without supervision, you know, they see this as, where's the parents? Yeah. So, so I call it self-supervised
learning because in fact, the underlying algorithms that are used are the same algorithms as the supervised
learning algorithms, except that what we trend them to do is not predict a particular set of
variables like the category of an image and not to predict a set of variables that have been
provided by human laborers. But what you're trying to machine to do is basically reconstruct
a piece of its input that is being masked out essentially.
You can think of it this way.
So show a piece of video automachine
and ask it to predict what's going to happen next.
And of course, after a while, you can
show what happens and the machine will kind of train
itself to do better at that task.
You can do all the latest, most successful models in natural language processing use self-supervised running.
You know, sort of bird style systems, for example, right?
You show it a window of a thousand words on a text corpus.
You take out 15% of the words, and then you train the machine to predict the words
that are missing that's super resonating. It's not predicting the future, it's just
predicting things in the middle, but you could have you predict the future, that's what
language models do. So you construct, so in an unsupervised way, you construct a
model of language. Do you think or or video, or the physical world,
or whatever, right?
How far do you think that can take us?
Do you think very far, I think,
understands anything?
To some level, it has a shadow understanding of text,
but it needs to, I mean, to have kind of true human level
intelligence that you need to ground language in reality.
So some people are attempting to do this, right?
Having systems that kind of have some visual representation of what is being talked about,
which is one reason you need those interactive environments, actually.
But this is like a huge technical problem that is not solved, and that explains why
self-supervisioning works in the context of natural language,
but does not work in the context, or at least not well, in the context of image
recognition and video, although it's making progress quickly.
And the reason that reason is the fact that
it's much easier to represent uncertainty in the prediction
in the context of natural language than it is in the context of things like video and images.
So, for example, if I ask you to predict what words I'm missing,
you know, 15% of the words that are taken out.
The possibility is a small.
That means small, right?
There is a hundred thousand words in the in the lexicon.
And what the machine splits out is a big probability vector, right?
It's a bunch of numbers between the and 1 that's 1 to 1.
And we know how to do this with computers.
So there representing uncertainty in the prediction is relatively easy.
And that's in my opinion why those techniques work for NLP.
For images, if you block a piece of an image and you have a system reconstructed a piece of the image, there are many possible answers.
There are all perfectly legit.
And how do you represent that the set of possible answers?
You can't train a system to make one prediction. You can't train a neural net to say, here it is. That's the image.
Because there's a whole set of things that are compatible with it.
So how do you get the machine to represent not a single output, but a whole set of outputs?
And, you know, similarly with video prediction, there's a lot of things that can happen
in the future video. You're looking at me right now. I'm not moving my head very much,
but, you know, I might, you know, turn my head to the left or to the right. Right. If you don't have a system that can predict this,
and you train it with least square
to kind of minimize the error with a prediction
and what I'm doing, what you get
is a blurry image of myself
in all possible future positions that I might be in,
which is not a good prediction.
But so there might be other ways
to do the self-supervision, right?
For visual scenes.
Like what?
If I knew I wouldn't tell you, I'd publish it first.
I don't know.
I know there might be.
So I mean, these are kind of,
there might be artificial ways of like self-play
in games, the way you can simulate part of the environment.
You can-
Oh, that doesn't solve the problem.
It's just a way of generating data.
But because you have more of a control,
like me, you can control, yeah, it's a way
to generate data.
And that's right.
And because you can do huge amounts of data generation,
that doesn't, you're right.
Well, it's a creeps up on the problem
from the side of data.
And you don't think that's the right way to creep up.
It doesn't solve this problem of handling uncertainty
in the world, right?
So if you have a machine learn a predictive model
of the world in a game that is deterministic
or quasi deterministic, it's easy, right?
Just give a few frames of the game to a confnet,
put a bunch of layers, and then have the game generate the next few frames.
And if the game is deterministic, it works fine.
And that includes feeding the system with the action that your little character is going to take.
The problem comes from the fact that the real world and most games are not entirely predictable.
So there you get those blurry predictions and you can't do planning with blurry predictions.
So if you have a perfect model of the world,
you can in your head run this model with a hypothesis for sequence of actions
and you're going to predict the outcome of that sequence of actions.
But if your model is imperfect, how can you plan? for a sequence of actions, and you're going to predict the outcome of that sequence of actions.
But if your model is imperfect, how can you plan? Yeah, I quickly explode. What are your thoughts on the extension of this, which topic I'm super
excited about? It's connected to something you're talking about in terms of robotics,
is active learning. So as opposed to sort of completely unsupervised self-supervised learning, you ask the system for human
help for selecting parts you want to annotate next. So if you think about a robot
exploring a space or a baby exploring a space or a system exploring a
data set, every once in a while asking for human input.
Do you see value in that kind of work?
I don't see transformative value.
It's going to make things that we can already do more efficient
or they will learn slightly more efficiently,
but it's not going to make machines
significantly more intelligent.
I think, and by the way, there is no opposition, there is no conflict between self-supervised learning, reinforcement learning, and supervised learning, or imitation learning, or active learning.
I see self-supervised learning as a preliminary to all of the above. Yes. So the example I use very often is how is it that so if you use classical
reinforcement learning, deep reinforcement learning if you want. The best methods today
so called model free reinforcement learning to learn to play Atari games, take about 80
hours of training to reach the level
that any human can reach in about 15 minutes.
They get better than humans, but it takes them a long time.
Alpha Star, OK?
The Oreo Vini House and his teams, the system
to play Starcraft plays a single map, a single type of player, and can reach
better than human level with about the equivalent of 200 years of training playing against itself.
It's 200 years, right? It's not something that no human can.
I mean, I'm not sure what else to take away from that.
Okay. Now, take those algorithms, the best our algorithms we have today,
to train a car to drive itself. It would probably have to drive millions of hours.
It will have to kill thousands of pedestrians. It will have to run into thousands of trees.
It will have to run off cliffs. and it had to run off cliffs multiple times before it figures out
that it's about our idea, first of all, and second of all, before it figures out how not to do it.
And so, I mean, this type of learning obviously does not reflect the kind of learning that animals
and humans do. There is something missing that's really, really important there. And my
hypothesis, which I've been advocating for like five years now, is that we have predictive
models of the world that include the ability to predict under uncertainty. And what allows
us to not run off a cliff when we learn to drive, most of us can learn to drive in about
20 or 30 hours
of training without ever crashing, causing any accident.
If we drive next to a cliff, we know that if we turn the
wheel to the right, the car is going to run off the cliff
and nothing good is going to come out of this.
Because we have a pretty good model of intuitive physics
that tells us the car is going to fall.
We know about gravity.
Babies run this around the age of eight or nine months.
That objects don't float the fault.
And we have a pretty good idea of the effect
of turning the wheel on the car,
and we need to stay on the road.
So there is a lot of things that we bring to the table,
which is basically our predictive model of the world.
And that model allows us to not do stupid things
and to basically stay within the context of things we need to do.
We still face unpredictable situations, and that's how we learn.
But that allows us to learn really, really, really quickly.
So that's called model-based reinforcement learning.
There's some imitation and super-vegening because we have
a driving instructor that tells us occasionally what to do.
But most of the learning is learning the model, learning physics that we've done since we were babies.
That's where almost all the learning physics is somewhat transferable from
it's transferable from sin to sin. Stupid things are the same everywhere. Yeah, I mean, if you, you know, you have an experience of the world, you don't need to be
particularly from a particular intelligence species to know that if you spill water from
a container, you know, the rest is going to get wet.
You might get wet.
So you know, cats know this, right?
So the main problem we need to solve is how do we learn
models of the world? That's, and that's what I'm interested in. That's what self-supervised
running is all about. If you were to try to construct a benchmark for, let's, let's look at
MNIST. I'd love that data set. But if you do think it's useful, interesting slash possible,
And if you do think it's useful, interesting slash possible to perform well on MNUS with just one example of each digit and how would we solve that problem?
The answer is probably yes.
The question is what other type of learning are you allowed to do?
So if what you're allowed to do is train on some gigantic dataset of labeled digit that's
called transfer learning. And we know that works. Okay. We do this
at Facebook like in production, right? We train large convolutional nets to
predict hashtags that people type on Instagram and we train on billions of
images, literally billions. And then we chop off the last layer and fine tune on
whatever task we want. That works really well. You can beat, you know, the
image network record with this.
We actually open-sourced the whole thing a few weeks ago.
Yeah, that's still pretty cool.
But yeah, so what in yet,
what would be impressive,
what's useful and impressive,
what kind of transfer learning would be useful and impressive?
Is it Wikipedia?
That kind of thing?
No, no, that's so-
I don't think transfer learning is really where we should focus.
We should try to do,
you know, have a kind of scenario for a benchmark where you have unlabeled data.
And you can, and it's very large number of unable data.
It could be video clips.
It could be where you do, you know, frame prediction.
It could be images where you could choose to you could choose to mask a piece of it,
could be whatever, but they're unlabeled
and you're not allowed to label them.
So you do some training on this,
and then you train on a particular supervised task,
ImageNet or Enlist,
and you measure how your test error or validation error
decreases as you increase a number of label training
samples.
And what you'd like to see is that your error decreases
much faster than if you train from scratch, from random
weights, so that to reach the same level of performance and
a completely supervised, purely supervised system,
would reach you would need way fewer samples.
So that's the crucial question,
because it will answer the question to people interested in medical image analysis.
If I want to get a particular level of error rate for this task,
I know I need a million samples.
Can I do, you know, self-supervised pre-training to reduce this to about 100 or something?
Anything they answer there is self-supervised pre-training.
Yeah. Some form of it.
Telling you active learning, but you disagree.
No, it's useless.
It's just not going to lead to a quantum leap.
It's just going to make things that we already do.
So your way smarter than me, I just disagree with you.
But I don't have anything to back that.
It's just intuition.
So I work with a lot of large scale data sets and there's something that might be magic
and active learning.
But okay. At least I said it publicly.
At least I'm being an idiot publicly. Okay. It's not being an idiot. It's working with the
data you have. I mean, certainly people are doing things like, okay, I have 3,000 hours of
imitation running for a start driving car, but most of those are incredibly boring. What I like is
select 10% of them that are kind of the most informative. are incredibly boring. What I like is select, you know, 10% of them
that are kind of the most informative. And with just that, I would probably reach the same.
So it's a weak form of active learning if you want. Yes, but there might be a much stronger version.
Yeah, that's right. That's what an awesome question exists. The question is how much
strong you can get. Elon Musk is is confident, talked to him recently.
He's confident that large-scale data in deep learning can solve the time of a driving
problem.
What are your thoughts on the limits, possibilities of deep learning in this space?
Of course, it's obviously part of the solution.
I mean, I don't think we'll ever have a set driving system, or it is not in the foreseeable
future, that does not use deep learning.
Let me put it this way.
Now, how much of it?
So in the history of engineering,
particularly AI-like systems,
there's generally a first phase where everything is built by hand,
and there is a second phase,
and that was the case for autonomous driving 20 you know, 20, 30 years ago.
There's a phase where there's a little bit of learning is used,
but there's a lot of engineering that's involved in kind of,
you know, taking care of corner cases and putting limits,
et cetera, because the learning system is not perfect.
And then as technology progresses,
we end up relying more and more on learning.
That's the history of character recognition,
so history of speech recognition,
not computer vision, natural language processing.
And I think the same is going to happen
with the time it's driving that currently
the methods that are closest to providing
some level of autonomy,
some decent level of autonomy,
where you don't expect a driver to kind of do anything,
is where you const't expect a driver to kind of do anything,
is where you constrain the world, so you only run within, you know, 100 square kilometers or square miles in Phoenix, but the weather is nice and the roads are wide, which is what WEMO is doing.
You completely over-engineer the car with tons of light hours and sophisticated sensors that
are too expensive for consumer cars,
but they're fine if you just run a fleet.
And you engineer the hell out of everything else,
you map the entire world,
so you have a complete 3D model of everything.
So the only thing that the perception system
has to take care of is moving objects and construction
and sort of things that weren't in your map.
And you can engineer a good slam system or a step, right?
So that's kind of the current approach that's closest to some level of autonomy.
But I think eventually the long-term solution is going to rely more and more on
learning and possibly using a combination of self-supervised learning and model-based
reinforcement or something like that.
But ultimately learning will be not just at the core, but really the fundamental part of the system.
Yeah, it already is, but it will become more and more.
What do you think it takes to build a system with human level intelligence?
You talked about the AI system and the Muir,
being way out of reach,
our current reach. This might be outdated as well, but this is your way out of reach. What would it
take to build her? Do you think? So I can tell you the first two obstacles that we have to clear,
but I don't know how many obstacles they are after this. So the image I usually use is that there
is a bunch of mountains that we have to climb. And we can see the first one, but we don't know how many of us took us there after this. So the image I usually use is that there is a bunch of mountains that we have to climb.
And we can see the first one.
But we don't know if there are 50 mountains behind the internet.
And this might be a good sort of metaphor
for why AI researchers in the past
have been overly optimistic about the result of a AI.
For example, New Orleans Simon wrote the general problem solver,
and they called it the general problem solver.
Yeah, problem solver.
Of course, the first thing you realize is that all the problems you want to solve are exponential.
You can't actually use it for anything useful.
But all you see is the first peak.
What are the first couple of peaks for her?
The first peak which is precisely what I'm working on,
is self-supervisioning.
How do we get machines to learn models of the world
by observation, kind of like babies and like young animals?
So we've been working with, you know,
cognitive scientists.
So this Amanda DuPou, who's at Fair in Paris,
is half time is also a researcher in French
University.
And he has his chart that shows that which how many months of life baby humans can learn
different concepts.
And you can measure this in various ways. So things like distinguishing animate objects
from any animate object, you can tell the difference
at age two, three months.
Whether an object is gonna stay stable,
it's gonna fall, you know,
about four months you can tell.
You know, there are various things like this.
And then things like gravity, the fact that objects
are not supposed to float in the air,
but are supposed to fall, you run this
around the edge of eight or nine months.
If you look at a lot of eight-month-old babies,
you give them a bunch of toys on their high chair.
First thing they do is they throw them on the ground
and they look at them.
It's because, you know, they're learning
about actively learning about gravity.
Graffiti, yeah.
So they're not trying to annoy you, but they need to do the experiment, right?
So how do we get machines to learn, like babies, mostly by observation, with a little
bit of interaction, and learning those models of the world?
Because I think that's really a crucial piece of an intelligent autonomous system.
So if you think about the architecture of an intelligent autonomous system, it needs to have a
predictive model of the world.
So something that says, here is a world at time T, here is a
state of the world at time T plus 1 if I take this action.
And it's not a single answer, it can be a contribution.
Yeah, well, we don't know how to represent
distributions in high-dimensional continuous spaces.
So it's got to be something weaker than that, okay?
But with some more presentation of uncertainty,
if you have that,
then you can do what optimal control theory is called
model predictive control,
which means that you can run your model
with a hypothesis for sequence of action
and then see the result.
Now, what you need, the other thing you need is some sort
of objective that you want to optimize.
Am I reaching the goal of grabbing this object? Am I minimizing energy? Am I whatever? Right? So there is some sort of
objectives that you have to minimize. And so in your head, if you had this model,
you can figure out the sequence of action that will optimize your objective.
That objective is something that ultimately is rooted in your bezel ganglia, at least in the
human brain. That's what it's a bezel ganglia,
computes your level of contentment or miscontentment.
I don't know if that's a word.
Unhappiness, okay?
Yeah, this contentment.
Discontentment.
Discontentment, mate.
And so your entire behavior is driven towards
kind of minimizing that objective,
which is maximizing your contentment,
computed by your bezogangliet.
And what you have is an objective function,
which is basically a predictor
of what your bezogangliet is gonna tell you.
So you're not gonna put your hand on fire
because you know it's gonna burn,
and you're gonna get hurt.
And you're predicting this
because of your model of the world,
and your predictor of the subjective.
So if you have those three components, you have the four components.
You have the hard-wired contentment objective computer if you want, calculator.
And then you have the three components. One is the objective predictor, which basically predicts your level contentment. One is the model of the world. And there's a third module
I didn't mention, which is a module that will figure out the best course of action to optimize
an objective given your model. Okay. Yeah.
Cool. It's a policy network or something like that, right?
Now, you need those three components
to act autonomously, intelligently.
And you can be stupid in three different ways.
You can be stupid because your model of the world is wrong.
You can be stupid because your objective
is not aligned with what you actually want to achieve.
Okay.
Inhumans that would be a psychopath.
And then the third thing you can be stupid
is that you have the right model.
You have the right objective.
But you're unable to figure out a course of action
to optimize your objective given your model.
OK.
Some people who are in charge of big countries
actually have all three that are wrong.
All right. Which country is that? I don't know. OK. Some people who are in charge of big countries actually have all three that are wrong.
All right.
Which countries?
I don't know.
Okay.
So, if we think about this agent, if you think about the movie, her, you've criticized
the art project that is Sophia the robot.
And what that project essentially does is uses our natural inclination
to anthropomorphize things that look like human and give them more.
Do you think that could be used by AI systems like in the movie Her?
So do you think that body is needed to create a feeling of intelligence?
Well, if Sophia was just on our piece,
I would have no problem with it,
but it's presented as something else.
Let me add in that comment real quick,
if creators of Sophia could change something
about their marketing or behavior in general,
what would it be?
What's, I'm just about everything.
I mean, don't you think, here's a tough question.
Let me, so I agree with you.
So Sophia is not, the general public feels that Sophia can do way more than she actually can.
That's right.
And the people who created Sophia are not honestly, publicly communicating
trying to teach the public.
But here's a tough question.
Don't you think the same thing is scientists
in industry and research are taking advantage
of the same misunderstanding in the public
when they create AI companies or publish stuff.
Some companies, yes. I mean, there is no sense of there's no desire to delude. There is no
desire to kind of over claim with something. Right. You can be sure paper on AI that has this
result on ImageNet. It's pretty clear. I mean, it's not even interesting anymore. But
I mean, it's not even interesting anymore, but I don't think there is that. I mean, the reviewers are generally not very forgiving of unsupported claims of this type.
But there are certainly quite a few startups that have had a huge amount of hype around
this that I find extremely damaging and have been calling it out when I've seen it.
So yeah, but to go back to your original question,
like the necessity of embodiment,
I think embodiment is necessary.
I think grounding is necessary.
So I don't think we're gonna get machines
that really understand language
without some level of grounding in the world.
And it's not clear to me that language
is a high enough bandwidth medium
to communicate how the real world works.
I think for this, we start to ground what grounding means.
So grounding means that, so there is this classic problem of common sense reasoning, you
know, the Wiener grad schema, right?
And so I tell you, the trophy doesn't fit in the suitcase because it's too big, or the
trophy doesn't fit in the suitcase because it's too big or the trophy doesn't fit in the suitcase because it's too small
And the it in the first case refers to the trophy in the second case to the suitcase and the reason you can figure this out
It's because you know the trophy in the suitcase are you know one is supposed to fit in the other one and you know the notion of size
And the big object doesn't fit in a small object unless it's a tardis you know things like that, right?
So you have this this knowledge of father world, you know, things like that, right? So you have this knowledge of how the world works,
of geometry and things like that.
I don't believe you can learn everything about the world,
but just being told in language how the world works.
I think you need some low-level perception of the world,
you know, be it visual touch, you know, whatever,
but some how you bend with perception of the world.
So by reading all the world's text, you still may not have enough information.
That's right.
There's a lot of things that just will never appear in text,
and that you can't really infer.
So I think common sense will emerge from, you know,
certainly a lot of language interaction,
but also with watching videos, or perhaps even interacting in virtual environments.
And possibly robot interacting in the real world. But I don't actually believe necessarily
that this last one is absolutely necessary. But I think there's a need for some grounding.
But the final product doesn't necessarily need to be embodied.
No, you're saying. It just needs to have an awareness grounding.
Right. But it needs to know how the world works to have, you know, to not be frustrated,
frustrating to talk to. And you talked about emotions being important. That's the whole
other topic. Well, so, you know, I talked about this, the, the Bezul Ganglia, Ganglia as the,
I talked about this, the Bezul Ganglia, Ganglia as the thing that calculates
your level of mixed-constantment contentment.
And then there is this other module
that sort of tries to do a prediction of
whether you're gonna be content or not.
That's the source of some emotion.
So fear, for example, is an anticipation
of bad things that can happen to you, right?
You have this inkling that there is some chance
that something really bad is going to happen to you
and that creates fear.
When you know for sure that something bad is going to happen
to you, you kind of give up, right?
It's not there anymore.
It's uncertainty that creates fear.
So the punchline is,
we're not going to have autonomous intelligence
without emotions.
Whatever the heck emotions are, do you mentioned very practical things of fear,
but there's a lot of other mess around it. But there are the results of drives. Yeah,
there's deeper biological stuff going on. And I've talked to a few folks on this. There's
fascinating stuff that ultimately connects to our brain. If we create an AGI system, human
level intelligence system, and you get to ask her one question, what would that question
be?
You know, I think the first one will create, will probably not be that smart. They'll be
like a four year old.
Okay. will probably not be that smart. They'll be like a four-year-old. So you would have to ask her a question
to know she's not that smart.
Yeah.
Well, what's a good question to ask,
to be as good as a person?
To be as good as a person.
To be as good as a person.
And if she answers, oh, it's because the leaves of the tree
are moving and they create a wind, She's onto something. And she says,
that's a stupid question. She's really onto something. No. And then you tell her, actually,
you know, here is the real thing. And she says, Oh, yeah, that makes sense. So questions that
reveal the ability to do common sense reasoning about the physical world. Yeah, and you'll summon up a little call to any of your friends.
Call to any of your friends. Well, it was a huge honor. Congratulations on your
turning award. Thank you so much for talking today. Thank you.