back to indexRegina Barzilay: Deep Learning for Cancer Diagnosis and Treatment | Lex Fridman Podcast #40
link |
The following is a conversation with Regina Barsley.
link |
She's a professor at MIT and a world class researcher
link |
in natural language processing and applications
link |
of deep learning to chemistry and oncology,
link |
or the use of deep learning for early diagnosis, prevention,
link |
and treatment of cancer.
link |
She has also been recognized for a teaching
link |
of several successful AI related courses at MIT,
link |
including the popular introduction to machine
link |
This is the Artificial Intelligence Podcast.
link |
If you enjoy it, subscribe on YouTube,
link |
give it 5,000 iTunes, support it on Patreon,
link |
or simply connect with me on Twitter
link |
at Lex Freedman, spelled F R I D M A N.
link |
And now here's my conversation with Regina Barsley.
link |
In an interview you've mentioned
link |
that if there's one course you would take,
link |
it would be a literature course with a friend of yours
link |
that a friend of yours teaches just out of curiosity
link |
because I couldn't find anything on it.
link |
Are there books or ideas that had profound impact
link |
on your life journey, books and ideas perhaps outside
link |
of computer science and the technical fields?
link |
I think because I'm spending a lot of my time at MIT
link |
and previously in other institutions where I was a student,
link |
I have a limited ability to interact with people.
link |
So a lot of what I know about the world
link |
actually comes from books.
link |
And there were quite a number of books
link |
that had profound impact on me and how I view the world.
link |
Let me just give you one example of such a book.
link |
I've maybe a year ago read a book
link |
called The Emperor of All Melodies.
link |
It's a book about, it's kind of a history of science book
link |
on how the treatments and drugs for cancer were developed.
link |
And that book, despite the fact that I am
link |
in the business of science, really opened my eyes
link |
on how imprecise and imperfect the discovery process is
link |
and how imperfect our current solutions.
link |
And what makes science succeed and be implemented
link |
and sometimes it's actually not the strengths of the idea
link |
but devotion of the person who wants to see it implemented.
link |
So this is one of the books that, you know,
link |
at least for the last year quite changed the way
link |
I'm thinking about scientific process
link |
just from the historical perspective.
link |
And what do I need to do to make my ideas really implemented?
link |
Let me give you an example of a book
link |
which is not kind of, which is a fiction book.
link |
It's a book called Americana.
link |
And this is a book about a young female student
link |
who comes from Africa to study in the United States.
link |
And it describes her past, you know, within her studies
link |
and her life transformation that, you know,
link |
in a new country and kind of adaptation to a new culture.
link |
And when I read this book,
link |
I saw myself in many different points of it.
link |
But it also kind of gave me the lens on different events
link |
and some events that I never actually paid attention
link |
to one of the funny stories in this book
link |
is how she arrives to her new college
link |
and she starts speaking in English
link |
and she had this beautiful British accent
link |
because that's how she was educated in her country.
link |
This is not my case.
link |
And then she notices that the person who talks to her,
link |
you know, talks to her in a very funny way,
link |
in a very slow way.
link |
And she's thinking that this woman is disabled
link |
and she's also trying to kind of to accommodate her.
link |
And then after a while,
link |
when she finishes her discussion with this officer
link |
from her college, she sees how she interacts
link |
with the other students, with American students
link |
and she discovers that actually she talked to her this way
link |
because she saw that she doesn't understand English.
link |
And he thought, wow, this is a funny experience.
link |
And literally within few weeks,
link |
I went to LA to a conference
link |
and they asked somebody in an airport, you know,
link |
how to find like a cab or something.
link |
And then I noticed that this person is talking
link |
in a very strange way.
link |
And my first thought was that this person
link |
have some, you know, pronunciation issues or something.
link |
And I'm trying to talk very slowly to him
link |
and I was with another professor, Ernst Frankel.
link |
And he's like laughing because it's funny
link |
that I don't get that the guy is talking in this way
link |
because he thinks that I cannot speak.
link |
So it was really kind of mirroring experience
link |
and it led me think a lot about my own experiences
link |
moving, you know, from different countries.
link |
So I think that books play a big role
link |
in my understanding of the world.
link |
On the science question, you mentioned that
link |
it made you discover that personalities
link |
of human beings are more important than perhaps ideas.
link |
Is that what I heard?
link |
It's not necessarily that they are more important
link |
than ideas, but I think that ideas on their own
link |
are not sufficient.
link |
And many times at least at the local horizon
link |
is the personalities and their devotion to their ideas
link |
is really that locally changes the landscape.
link |
Now, if you're looking at AI, like let's say 30 years ago,
link |
you know, dark ages of AI or whatever,
link |
what the symbolic times you can use any word,
link |
you know, there were some people,
link |
now we are looking at a lot of that work
link |
and we are kind of thinking this was not really
link |
maybe a relevant work.
link |
But you can see that some people managed to take it
link |
and to make it so shiny and dominate the, you know,
link |
the academic world and make it to be the standard.
link |
If you look at the area of natural language processing,
link |
it is well known fact that the reason the statistics
link |
in NLP took such a long time to become mainstream
link |
because there were quite a number of personalities
link |
which didn't believe in this idea
link |
and then stop research progress in this area.
link |
So I do not think that, you know, kind of asymptotically
link |
maybe personalities matters,
link |
but I think locally it does make quite a bit of impact
link |
and it's generally, you know, speed up,
link |
speeds up the rate of adoption of the new ideas.
link |
Yeah, and the other interesting question is in the early days
link |
of particular discipline, I think you mentioned in that book
link |
was, is ultimately a book of cancer?
link |
It's called The Emperor of All Melodies.
link |
Yeah, and those melodies included the trying to,
link |
the medicine, was it centered around?
link |
So it was actually centered on, you know,
link |
how people thought of curing cancer.
link |
Like for me, it was really a discovery how people,
link |
what was the science of chemistry behind drug development
link |
that it actually grew up out of dyeing,
link |
like coloring industry that people who develop chemistry
link |
in 19th century in Germany and Britain to do,
link |
you know, the really new dyes,
link |
they looked at the molecules and identified
link |
that they do certain things to cells.
link |
And from there, the process started and, you know,
link |
like historians think, yeah, this is fascinating
link |
that they managed to make the connection
link |
and look under the microscope and do all this discovery.
link |
But as you continue reading about it
link |
and you read about how chemotherapy drugs
link |
were actually developed in Boston
link |
and some of them were developed
link |
and Dr. Farber from Dana Farber,
link |
you know, how the experiments were done,
link |
that, you know, there was some miscalculation.
link |
Let's put it this way.
link |
And they tried it on the patients
link |
and they just, and those were children with leukemia
link |
and they died and they tried another modification.
link |
You look at the process, how imperfect is this process?
link |
And, you know, like, if we're again looking back
link |
like 60 years ago, 70 years ago,
link |
you can kind of understand it.
link |
But some of the stories in this book,
link |
which were really shocking to me,
link |
were really happening, you know, maybe decades ago.
link |
And we still don't have a vehicle
link |
to do it much more fast and effective and, you know,
link |
scientific the way I'm thinking,
link |
computer science scientific.
link |
So from the perspective of computer science,
link |
you've got a chance to work the application to cancer
link |
and to medicine in general.
link |
From a perspective of an engineer and a computer scientist,
link |
how far along are we from understanding the human body,
link |
biology, of being able to manipulate it in a way
link |
we can cure some of the maladies, some of the diseases?
link |
So this is a very interesting question.
link |
And if you're thinking as a computer scientist
link |
about this problem, I think one of the reasons
link |
that we succeeded in the areas
link |
we as a computer scientist succeeded
link |
is because we don't have,
link |
we are not trying to understand in some ways.
link |
Like if you're thinking about like eCommerce, Amazon,
link |
Amazon doesn't really understand you.
link |
And that's why it recommends you certain books
link |
or certain products, correct?
link |
And in, you know, traditionally,
link |
when people were thinking about marketing, you know,
link |
they divided the population
link |
to different kind of subgroups,
link |
identify the features of the subgroup
link |
and come up with a strategy
link |
which is specific to that subgroup.
link |
If you're looking about recommendations,
link |
they're not claiming that they're understanding somebody,
link |
they're just managing from the patterns of your behavior
link |
to recommend you a product.
link |
Now, if you look at the traditional biology,
link |
obviously I wouldn't say that I am at any way,
link |
you know, educated in this field.
link |
But, you know, what I see,
link |
there is really a lot of emphasis
link |
on mechanistic understanding.
link |
And it was very surprising to me coming from computer science
link |
how much emphasis is on this understanding.
link |
And given the complexity of the system,
link |
maybe the deterministic full understanding
link |
of this process is, you know, beyond our capacity.
link |
And the same way as in computer science,
link |
when we're doing recognition,
link |
when we're doing recommendation in many other areas,
link |
it's just probabilistic matching process.
link |
And in some way, maybe in certain cases,
link |
we shouldn't even attempt to understand
link |
or we can attempt to understand, but in parallel,
link |
we can actually do this kind of matching
link |
that would help us to find Qo to do early diagnostics
link |
And I know that in these communities,
link |
it's really important to understand,
link |
but I am sometimes wondering,
link |
what exactly does it mean to understand here?
link |
Well, there's stuff that works,
link |
and, but that can be, like you said,
link |
separate from this deep human desire
link |
to uncover the mysteries of the universe,
link |
of science, of the way the body works,
link |
the way the mind works.
link |
It's the dream of symbolic AI,
link |
of being able to reduce human knowledge into logic
link |
and be able to play with that logic
link |
in a way that's very explainable
link |
and understandable for us humans.
link |
I mean, that's a beautiful dream.
link |
So I understand it, but it seems that
link |
what seems to work today, and we'll talk about it more,
link |
is as much as possible, reduce stuff into data,
link |
reduce whatever problem you're interested in to data
link |
and try to apply statistical methods,
link |
apply machine learning to that.
link |
On a personal note,
link |
you were diagnosed with breast cancer in 2014.
link |
Would it facing your mortality make you think about
link |
how did it change you?
link |
You know, this is a great question.
link |
And I think that I was interviewed many times
link |
and nobody actually asked me this question.
link |
I think I was 43 at a time.
link |
And the first time I realized in my life that I may die.
link |
And I never thought about it before.
link |
And yeah, and there was a long time
link |
since you diagnosed until you actually know
link |
what you have and how severe is your disease.
link |
For me, it was like maybe two and a half months.
link |
And I didn't know where I am during this time
link |
because I was getting different tests.
link |
And one would say, it's bad.
link |
And I would say, no, it is not.
link |
So until I knew where I am,
link |
I really was thinking about
link |
all these different possible outcomes.
link |
Were you imagining the worst?
link |
Or were you trying to be optimistic?
link |
It would be really, I don't remember, you know,
link |
what was my thinking?
link |
It was really a mixture with many components at the time
link |
at speaking, you know, in our terms.
link |
And one thing that I remember,
link |
and you know, every test comes and then you think,
link |
oh, it could be this or it may not be this.
link |
And you're hopeful and then you're desperate.
link |
So it's like, there is a whole, you know,
link |
slew of emotions that goes through you.
link |
But what I remember is that when I came back to MIT,
link |
I was kind of going the whole time
link |
through the treatment to MIT,
link |
but my brain was not really there.
link |
But when I came back really, finished my treatment
link |
and I was here teaching and everything.
link |
You know, I look back at what my group was doing,
link |
what other groups was doing,
link |
and I saw these three realities.
link |
It's like people are building their careers
link |
on improving some parts around two or three percent
link |
I was, it's like, seriously,
link |
I did a work on how to decipher eukaryotic,
link |
like a language that nobody speak and whatever,
link |
like what is significance?
link |
When I was sad, you know, I walked out of MIT,
link |
which is, you know, when people really do care,
link |
you know, what happened to your iClear paper,
link |
you know, what is your next publication,
link |
to ACL, to the world where people, you know,
link |
people, you see a lot of suffering
link |
that I'm kind of totally shielded on it on a daily basis.
link |
And it's like the first time I've seen like real life
link |
and real suffering.
link |
And I was thinking, why are we trying to improve the parser
link |
or deal with some trivialities
link |
when we have capacity to really make a change?
link |
And it was really challenging to me
link |
because on one hand, you know,
link |
I have my graduate students who really want to do
link |
their papers and their work
link |
and they want to continue to do what they were doing,
link |
And then it was me who really kind of reevaluated
link |
what is importance and also at that point
link |
because I had to take some break.
link |
I look back into like my years in science
link |
and I was thinking, you know,
link |
like 10 years ago, this was the biggest thing.
link |
I don't know, topic models.
link |
We have like millions of papers on topic models
link |
and variation of topics models now.
link |
It's totally like irrelevant.
link |
And you start looking at this, you know,
link |
what do you perceive as important
link |
at different point of time
link |
and how, you know, it fades over time.
link |
And since we have a limited time,
link |
all of us have limited time on us.
link |
It's really important to prioritize things
link |
that really matter to you,
link |
maybe matter to you at that particular point,
link |
but it's important to take some time
link |
and understand what matters to you,
link |
which may not necessarily be the same
link |
as what matters to the rest of your scientific community
link |
and pursue that vision.
link |
And so though that moment, did it make you cognizant?
link |
You mentioned suffering of just the general amount
link |
of suffering in the world.
link |
Is that what you're referring to?
link |
So as opposed to topic models and specific detail problems
link |
in NLP, did you start to think about other people
link |
who have been diagnosed with cancer?
link |
Is that the way you saw the,
link |
started to see the world perhaps?
link |
And it actually creates because like, for instance,
link |
you know, there's parts of the treatment
link |
where you need to go to the hospital every day
link |
and you see, you know, the community of people that you see
link |
and many of them are much worse than I was at a time
link |
and you're all of a sudden see it all.
link |
And people who are happier someday
link |
just because they feel better.
link |
And for people who are in our normal realm,
link |
you take it totally for granted that you feel well,
link |
that if you decide to go running, you can go running
link |
and you can, you know, you're pretty much free to do
link |
whatever you want with your body.
link |
Like I saw like a community,
link |
my community became those people.
link |
And I remember one of my friends,
link |
Dina Katabi took me to Prudential
link |
to buy me a gift for my birthday.
link |
And it was like the first time in months
link |
that I went to kind of to see other people.
link |
And I was like, wow.
link |
First of all, these people, you know,
link |
they're happy and they're laughing
link |
and they're very different from this other my people.
link |
And second of thing, are they totally crazy?
link |
They're like laughing and wasting their money
link |
on some stupid gifts.
link |
And, you know, they may die.
link |
They already may have cancer
link |
and they don't understand it.
link |
So you can really see how the mind changes
link |
that you can see that, you know,
link |
before that you can ask,
link |
didn't you know that you're gonna die?
link |
Of course I knew, but it was kind of a theoretical notion.
link |
It wasn't something which was concrete.
link |
And at that point when you really see it
link |
and see how little means sometime the system
link |
has to help them, you really feel
link |
that we need to take a lot of our brilliance
link |
that we have here at MIT
link |
and translate it into something useful.
link |
And you still couldn't have a lot of definitions,
link |
but of course alleviating, suffering, alleviating,
link |
trying to cure cancer is a beautiful mission.
link |
So I, of course, know the theoretically the notion
link |
of cancer, but just reading more and more
link |
about it's the 1.7 million new cancer cases
link |
in the United States every year,
link |
600,000 cancer related deaths every year.
link |
So this has a huge impact, United States globally.
link |
When broadly, before we talk about how machine learning,
link |
how MIT can help, when do you think
link |
we as a civilization will cure cancer?
link |
How hard of a problem is it from everything
link |
you've learned from it recently?
link |
I cannot really assess it.
link |
What I do believe will happen with the advancement
link |
in machine learning that a lot of types of cancer
link |
we will be able to predict way early
link |
and more effectively utilize existing treatments.
link |
I think, I hope at least that with all the advancements
link |
in AI and drug discovery, we would be able
link |
to much faster find relevant molecules.
link |
What I'm not sure about is how long it will take
link |
the medical establishment and regulatory bodies
link |
to kind of catch up and to implement it.
link |
And I think this is a very big piece of puzzle
link |
that is currently not addressed.
link |
That's the really interesting question.
link |
So first, a small detail that I think the answer is yes,
link |
but is cancer one of the diseases
link |
that when detected earlier,
link |
that's a significantly improves the outcomes.
link |
Cause we will talk about, there's the cure
link |
and then there is detection.
link |
And I think one machine learning can really help
link |
is earlier detection.
link |
So does detection help?
link |
Detection is crucial.
link |
For instance, the vast majority of pancreatic cancer patients
link |
are detected at the stage that they are incurable.
link |
That's why they have such a terrible survival rate.
link |
It's like just a few percent over five years.
link |
It's pretty much today a death sentence.
link |
But if you can discover this disease early,
link |
there are mechanisms to treat it.
link |
And in fact, I know a number of people who were diagnosed
link |
and saved just because they had food poisoning.
link |
They had terrible food poisoning.
link |
They went to ER, they got scan.
link |
There were early signs on the scan
link |
and that would save their lives.
link |
But this wasn't really an accidental case.
link |
So as we become better,
link |
we would be able to help too many more people
link |
that are likely to develop diseases.
link |
And I just want to say that as I got more into this field,
link |
I realized that cancer is of course a terrible disease
link |
when there are really the whole slew
link |
of terrible diseases out there,
link |
like neurodegenerative diseases and others.
link |
So we, of course, a lot of us are fixated on cancer
link |
just because it's so prevalent in our society.
link |
And you see these people when there are a lot of patients
link |
with neurodegenerative diseases
link |
and the kind of aging diseases
link |
that we still don't have a good solution for.
link |
And I felt as a computer scientist,
link |
we kind of decided that it's other people's job
link |
to treat these diseases
link |
because it's like traditionally people in biology
link |
or in chemistry or MDs are the ones who are thinking about it.
link |
And after kind of start paying attention,
link |
I think that it's really a wrong assumption
link |
and we all need to join the battle.
link |
So how it seems like in cancer specifically
link |
that there's a lot of ways that machine learning can help.
link |
So what's the role of machine learning
link |
and machine learning in the diagnosis of cancer?
link |
So for many cancers today,
link |
we really don't know what is your likelihood to get cancer.
link |
And for the vast majority of patients,
link |
especially on the younger patients,
link |
it really comes as a surprise.
link |
Like for instance, for breast cancer,
link |
80% of the patients are first in their families,
link |
And I never saw that I had any increased risk
link |
because nobody had it in my family
link |
and for some reason in my head,
link |
it was kind of inherited disease.
link |
But even if I would pay attention,
link |
the models that currently,
link |
there's very simplistic statistical models
link |
that are currently used in clinical practice
link |
that really don't give you an answer, so you don't know.
link |
And the same true for pancreatic cancer,
link |
the same true for non smoking lung cancer and many others.
link |
So what machine learning can do here
link |
is utilize all this data to tell us Ellie,
link |
who is likely to be susceptible
link |
and using all the information that is already there,
link |
be it imaging, be it your other tests
link |
and eventually liquid biopsies and others,
link |
where the signal itself is not sufficiently strong
link |
for human eye to do good discrimination
link |
because the signal may be weak,
link |
but by combining many sources,
link |
machine which is trained on large volumes of data
link |
can really detect it Ellie
link |
and that's what we've seen with breast cancer
link |
and people are reporting it in other diseases as well.
link |
That really boils down to data, right?
link |
And in the different kinds of sources of data.
link |
And you mentioned regulatory challenges.
link |
So what are the challenges
link |
in gathering large data sets in the space?
link |
Again, another great question.
link |
So it took me after I decided
link |
that I want to work on it two years to get access to data.
link |
And you did, like any significant amount.
link |
Like right now in this country,
link |
there is no publicly available data set of modern mammograms
link |
that you can just go on your computer,
link |
sign a document and get it.
link |
It just doesn't exist.
link |
I mean, obviously every hospital has its own collection
link |
of mammograms, there are data that came out
link |
of clinical trials.
link |
What we're talking about here is a computer scientist
link |
who just want to run his or her model
link |
and see how it works.
link |
This data, like ImageNet, doesn't exist.
link |
And there is an set which is called like Florid data set
link |
which is a film mammogram from 90s
link |
which is totally not representative
link |
of the current developments,
link |
whatever you're learning on them doesn't scale up.
link |
This is the only resource that is available.
link |
And today there are many agencies that govern access to data,
link |
like the hospital holds your data
link |
and the hospital decides whether they would give it
link |
to the researcher to work with this data or not.
link |
In the individual hospital?
link |
Yeah, I mean, the hospital may, you know,
link |
assuming that you're doing research collaboration,
link |
you can submit, you know,
link |
there is a proper approval process guided by IRB
link |
and if you go through all the processes,
link |
you can eventually get access to the data.
link |
But if you yourself know our AI community,
link |
there are not that many people
link |
who actually ever got access to data
link |
because it's very challenging process.
link |
And sorry, just in a quick comment,
link |
MGH or any kind of hospital,
link |
are they scanning the data?
link |
Are they digitally storing it?
link |
Oh, it is already digitally stored.
link |
You don't need to do any extra processing steps.
link |
It's already there in the right format.
link |
Is that right now there are a lot of issues
link |
that govern access to the data
link |
because the hospital is legally responsible for the data.
link |
And, you know, they have a lot to lose
link |
if they give the data to the wrong person,
link |
but they may not have a lot to gain
link |
if they give it as a hospital, as a legal entity
link |
as given it to you.
link |
And the way, you know, what I would mention
link |
happening in the future is the same thing
link |
that happens when you're getting your driving license.
link |
You can decide whether you want to donate your organs.
link |
So you can imagine that whenever a person
link |
goes to the hospital,
link |
it should be easy for them to donate their data for research
link |
and it can be different kind of,
link |
do they only give you a test results
link |
or only imaging data or the whole medical record?
link |
Because at the end,
link |
we all will benefit from all this insights.
link |
And it's only gonna say, I want to keep my data private,
link |
but I would really love to get it from other people
link |
because other people are thinking the same way.
link |
So if there is a mechanism to do this donation
link |
and the patient has an ability to say
link |
how they want to use their data for research,
link |
it would be really a game changer.
link |
People, when they think about this problem,
link |
there's a depends on the population,
link |
depends on the demographics,
link |
but there's some privacy concerns.
link |
Generally, not just medical data,
link |
just any kind of data.
link |
It's what you said, my data,
link |
it should belong kinda to me,
link |
I'm worried how it's gonna be misused.
link |
How do we alleviate those concerns?
link |
Because that seems like a problem that needs to be,
link |
that problem of trust, of transparency needs to be solved
link |
before we build large data sets that help detect cancer,
link |
help save those very people in the future.
link |
So seeing that two things that could be done,
link |
there is a technical solutions
link |
and there are societal solutions.
link |
So on the technical end,
link |
we today have ability to improve disambiguation,
link |
like for instance, for imaging,
link |
it's, you know, for imaging, you can do it pretty well.
link |
What's disambiguation?
link |
And disambiguation, sorry, disambiguation,
link |
removing the identification,
link |
removing the names of the people.
link |
There are other data, like if it is a raw text,
link |
you cannot really achieve 99.9%
link |
but there are all these techniques
link |
that actually some of them are developed at MIT,
link |
how you can do learning on the encoded data
link |
where you locally encode the image,
link |
you train on network,
link |
which only works on the encoded images
link |
and then you send the outcome back to the hospital
link |
and you can open it up.
link |
So those are the technical solutions.
link |
There are a lot of people who are working in this space
link |
where the learning happens in the encoded form.
link |
We are still early,
link |
but this is an interesting research area
link |
where I think we'll make more progress.
link |
There is a lot of work in natural language processing
link |
community, how to do the identification better.
link |
But even today, there are already a lot of data
link |
which can be identified perfectly,
link |
like your test data, for instance, correct,
link |
where you can just, you know,
link |
the name of the patient,
link |
you just want to extract the part with the numbers.
link |
The big problem here is again,
link |
hospitals don't see much incentive
link |
to give this data away on one hand
link |
and then there is general concern.
link |
Now, when I'm talking about societal benefits
link |
and about the education,
link |
the public needs to understand
link |
and I think that there are situations
link |
that I still remember myself
link |
when I really needed an answer.
link |
I had to make a choice
link |
and there was no information to make a choice.
link |
You're just guessing.
link |
And at that moment,
link |
you feel that your life is at the stake,
link |
but you just don't have information to make the choice.
link |
And many times when I give talks,
link |
I get emails from women who say,
link |
you know, I'm in this situation,
link |
can you please run statistic
link |
and see what are the outcomes?
link |
We get almost every week a mammogram
link |
that comes by mail to my office at MIT.
link |
I'm serious that people ask to run
link |
because they need to make, you know,
link |
life changing decisions.
link |
And of course, you know,
link |
I'm not planning to open a clinic here,
link |
but we do run and give them the results for their doctors.
link |
But the point that I'm trying to make
link |
that we all at some point or our loved ones
link |
will be in the situation where you need information
link |
to make the best choice.
link |
And if this information is not available,
link |
you would feel vulnerable and unprotected.
link |
And then the question is, you know,
link |
what do I care more?
link |
Because at the end everything is a trade off, correct?
link |
Just out of curiosity,
link |
what it seems like one possible solution,
link |
I'd like to see what you think of it
link |
based on what you just said,
link |
based on wanting to know answers
link |
for when you're yourself in that situation.
link |
Is it possible for patients to own their data
link |
as opposed to hospitals owning their data?
link |
Of course, theoretically,
link |
I guess patients own their data,
link |
but can you walk out there with a USB stick
link |
containing everything or upload it to the cloud
link |
where a company, you know,
link |
I remember Microsoft had a service,
link |
like I try, I was really excited about
link |
and Google Health was there.
link |
I tried to give, I was excited about it.
link |
Basically companies helping you upload your data
link |
to the cloud so that you can move from hospital to hospital
link |
from doctor to doctor.
link |
Do you see a promise of that kind of possibility?
link |
I absolutely think this is, you know,
link |
the right way to exchange the data.
link |
I don't know now who's the biggest player in this field,
link |
but I can clearly see that even for totally selfish health
link |
reasons, when you are going to a new facility
link |
and many of us are sent to some specialized treatment,
link |
they don't easily have access to your data.
link |
And today, you know, we would want to send us
link |
Mammogram need to go to their hospital,
link |
find some small office,
link |
which gives them the CD and they ship as a CD.
link |
So you can imagine we're looking at kind of decades old
link |
mechanism of data exchange.
link |
So I definitely think this is an area where hopefully
link |
all the right regulatory and technical forces will align
link |
and we will see it actually implemented.
link |
It's sad because unfortunately,
link |
and I have, I need to research why that happened,
link |
but I'm pretty sure Google Health and Microsoft Health Vault
link |
or whatever it's called, both closed down,
link |
which means that there was either regulatory pressure
link |
or there's not a business case
link |
or there's challenges from hospitals,
link |
which is very disappointing.
link |
So when you say, you don't know what the biggest players are,
link |
the two biggest that I was aware of closed their doors.
link |
So I'm hoping I'd love to see why
link |
and I'd love to see who else can come up.
link |
It seems like one of those Elon Musk style problems
link |
that are obvious needs to be solved
link |
and somebody needs to step up
link |
and actually do this large scale data collection.
link |
So I know there is an initiative in Massachusetts,
link |
a thing actually led by the governor
link |
to try to create this kind of health exchange system
link |
where at least to help people who are kind of when you show up
link |
in emergency room and there is no information
link |
about what are your allergies and other things.
link |
So I don't know how far it will go,
link |
but another thing that you said and I find it very interesting
link |
is actually who are the successful players in this space
link |
and the whole implementation.
link |
To me, it is from the anthropological perspective,
link |
it's more fascinating that AI that today goes in health care.
link |
We've seen so many attempts and so very little successes
link |
and it's interesting to understand that I by no means
link |
have knowledge to assess why we are in the position
link |
Yeah, it's interesting because data is really fuel
link |
for a lot of successful applications
link |
and when that data requires regulatory approval
link |
like the FDA or any kind of approval,
link |
it seems that the computer scientists are not quite there yet
link |
in being able to play the regulatory game,
link |
understanding the fundamentals of it.
link |
I think that in many cases when even people do have data,
link |
we still don't know what exactly do you need to demonstrate
link |
to change the standard of care.
link |
Let me give you an example related to my breast cancer research.
link |
So in traditional breast cancer risk assessment,
link |
there is something called density
link |
which determines the likelihood of a woman to get cancer
link |
and this is pretty much says how much white
link |
do you see on the mammogram?
link |
The whiter it is, the more likely the tissue is dense.
link |
And the idea behind density,
link |
it's not a bad idea,
link |
in 1967 a radiologist called Wolf decided to look back
link |
at women who were diagnosed
link |
and see what is special in their images.
link |
Can we look back and say that they're likely to develop?
link |
So he come up with some patterns
link |
and it was the best that his human eye can identify
link |
then it was kind of formalized and coded into four categories
link |
and that's what we are using today.
link |
And today this density assessment is actually a federal law
link |
from 2019 approved by President Trump
link |
and for the previous FDA commissioner
link |
where women are supposed to be advised by their providers
link |
if they have high density,
link |
putting them into higher risk category
link |
and in some states you can actually get supplementary screening
link |
paid by your insurance because you are in this category.
link |
Now you can say how much science do we have behind it?
link |
Whatever biological science or epidemiological evidence.
link |
So it turns out that between 40 and 50% of women
link |
have dense breast.
link |
So above 40% of patients are coming out of their screening
link |
and somebody tells them you are in high risk.
link |
Now what exactly does it mean
link |
if you as half of the population in high risk
link |
gets from saying maybe I'm not,
link |
or what do I really need to do with it?
link |
Because the system doesn't provide me a lot of the solutions
link |
because there are so many people like me,
link |
we cannot really provide very expensive solutions for them.
link |
And the reason this whole density became this big deal
link |
it's actually advocated by the patients
link |
who felt very unprotected because many women
link |
when did the mammograms which were normal
link |
and then it turns out that they already had cancer,
link |
quite developed cancer.
link |
So they didn't have a way to know who is really at risk
link |
and what is the likelihood that when the doctor tells you
link |
you're okay, you are not okay.
link |
So at the time and it was 15 years ago,
link |
this maybe was the best piece of science that we had
link |
and it took quite 15, 16 years to make it federal law.
link |
But now this is a standard.
link |
Now with a deep learning model
link |
we can so much more accurately predict
link |
who is gonna develop breast cancer
link |
just because you're trained on a logical thing.
link |
And instead of describing how much white
link |
and what kind of white machine
link |
can systematically identify the patterns
link |
which was the original idea behind the sort
link |
of the tradiologist,
link |
machines can do it much more systematically
link |
and predict the risk when you're training the machine
link |
to look at the image and to say the risk in one to five years.
link |
Now you can ask me how long it will take
link |
to substitute this density
link |
which is broadly used across the country
link |
and really it's not helping to bring this new models.
link |
And I would say it's not a matter of the algorithm.
link |
Algorithm is already orders of magnitude better
link |
than what is currently in practice.
link |
I think it's really the question,
link |
who do you need to convince?
link |
How many hospitals do you need to run the experiment?
link |
What, you know, all this mechanism of adoption
link |
and how do you explain to patients
link |
and to women across the country
link |
that this is really a better measure?
link |
And again, I don't think it's an AI question.
link |
We can walk more and make the algorithm even better
link |
but I don't think that this is the current, you know,
link |
the barrier, the barrier is really this other piece
link |
that for some reason is not really explored.
link |
It's like anthropological piece.
link |
And coming back to your question about books,
link |
there is a book that I'm reading.
link |
It's called American Sickness by Elizabeth Rosenthal
link |
and I got this book from my clinical collaborator,
link |
And I said, I know everything that I need to know
link |
about American health system,
link |
but you know, every page doesn't fail to surprise me.
link |
And I think that there is a lot of interesting
link |
and really deep lessons for people like us
link |
from computer science who are coming into this field
link |
to really understand how complex is the system of incentives
link |
in the system to understand how you really need
link |
to play to drive adoption.
link |
You just said it's complex,
link |
but if we're trying to simplify it,
link |
who do you think most likely would be successful
link |
if we push on this group of people?
link |
Is it the doctors?
link |
Is it the hospitals?
link |
Is it the governments or policy makers?
link |
Is it the individual patients, consumers
link |
who needs to be inspired to most likely lead to adoption?
link |
Or is there no simple answer?
link |
There's no simple answer,
link |
but I think there is a lot of good people in medical system
link |
who do want to make a change.
link |
And I think a lot of power will come from us as a consumers
link |
because we all are consumers or future consumers
link |
of healthcare services.
link |
And I think we can do so much more
link |
in explaining the potential and not in the hype terms
link |
and not saying that we're now cured or Alzheimer
link |
and I'm really sick of reading this kind of articles
link |
which make these claims.
link |
But really to show with some examples
link |
what this implementation does
link |
and how it changes the care.
link |
Because I can't imagine,
link |
it doesn't matter what kind of politician it is,
link |
we all are susceptible to these diseases.
link |
There is no one who is free.
link |
And eventually, we all are humans
link |
and we are looking for a way to alleviate the suffering.
link |
And this is one possible way
link |
where we currently are underutilizing,
link |
which I think can help.
link |
So it sounds like the biggest problems are outside of AI
link |
in terms of the biggest impact at this point.
link |
But are there any open problems
link |
in the application of ML to oncology in general?
link |
So improving the detection
link |
or any other creative methods,
link |
whether it's on the detection segmentations
link |
or the vision perception side
link |
or some other clever of inference.
link |
Yeah, what in general in your view
link |
are the open problems in this space?
link |
So I just want to mention that beside detection,
link |
another area where I am kind of quite active
link |
and I think it's really an increasingly important area
link |
in healthcare is drug design.
link |
Because it's fine if you detect something early,
link |
but you still need to get drugs
link |
and new drugs for these conditions.
link |
And today, all of the drug design, ML is non existent there.
link |
We don't have any drug that was developed by the ML model
link |
or even not developed,
link |
but at least even you,
link |
that ML model plays some significant role.
link |
I think this area with all the new ability
link |
to generate molecules with desired properties
link |
to do in silica screening is really a big open area.
link |
It to be totally honest with you,
link |
when we are doing diagnostics and imaging,
link |
primarily taking the ideas that were developed
link |
for other areas and you're applying them with some adaptation.
link |
The area of drug design
link |
is really technically interesting and exciting area.
link |
You need to work a lot with graphs
link |
and capture various 3D properties.
link |
There are lots and lots of opportunities
link |
to be technically creative.
link |
And I think there are a lot of open questions in this area.
link |
We're already getting a lot of successes
link |
even with the kind of the first generation of this models,
link |
but there is much more new creative things that you can do.
link |
And what's very nice to see is actually the more powerful,
link |
the more interesting models actually do better.
link |
So there is a place to innovate in machine learning
link |
And some of these techniques are really unique too,
link |
let's say to graph generation and other things.
link |
What just to interrupt really quick, I'm sorry.
link |
Graph generation or graphs, drug discovery in general.
link |
How do you discover a drug?
link |
Is this chemistry?
link |
Is this trying to predict different chemical reactions?
link |
Or is it some kind of...
link |
What do graphs even represent in this space?
link |
And what's a drug?
link |
Okay, so let's say you think that there are many different
link |
types of drugs, but let's say you're going to talk
link |
about small molecules because I think today,
link |
the majority of drugs are small molecules.
link |
So small molecule is a graph.
link |
The molecule is just where the node in the graph is an atom
link |
and then you have the bond.
link |
So it's really a graph representation
link |
if you look at it in 2D, correct?
link |
You can do it 3D, but let's say, well,
link |
let's keep it simple and stick in 2D.
link |
So pretty much my understanding today,
link |
how it is done at scale in the companies,
link |
you're without machine learning,
link |
you have high throughput screening.
link |
So you know that you are interested to get certain
link |
biological activity of the compounds.
link |
So you scan a lot of compounds,
link |
like maybe hundreds of thousands,
link |
some really big number of compounds.
link |
You identify some compounds which have the right activity
link |
and then at this point, the chemists come
link |
and they're trying to now to optimize this original heat
link |
to different properties that you want it to be,
link |
maybe soluble, you want to decrease toxicity,
link |
you want to decrease the side effects.
link |
Are those, sorry, again to the drop,
link |
can that be done in simulation
link |
or just by looking at the molecules
link |
or do you need to actually run reactions
link |
in real labs with lab posts and stuff?
link |
So when you do high throughput screening,
link |
you really do screening, it's in the lab.
link |
It's really the lab screening,
link |
you screen the molecules, correct?
link |
I don't know what screening is.
link |
The screening, you just check them for certain property.
link |
Like in the physical space, in the physical world,
link |
like actually there's a machine probably
link |
that's actually running the reaction.
link |
Actually running the reactions, yeah.
link |
So there is a process where you can run
link |
and that's why it's called high throughput,
link |
you know, it becomes cheaper and faster
link |
to do it on very big number of molecules.
link |
You run the screening, you identify potential,
link |
you know, potential good starts
link |
and then where the chemists come in
link |
who, you know, have done it many times
link |
and then they can try to look at it
link |
and say, how can you change the molecule
link |
to get the desired profile in terms of all other properties?
link |
So maybe how do I make it more bioactive and so on?
link |
And there, you know, the creativity of the chemists
link |
really is the one that determines the success
link |
of this design because again,
link |
they have a lot of domain knowledge of, you know,
link |
what works, how do you decrease the CCT and so on?
link |
And that's what they do.
link |
So all the drugs that are currently, you know,
link |
in the FDA approved drugs or even drugs
link |
that are in clinical trials,
link |
they are designed using these domain experts
link |
which goes through this combinatorial space
link |
of molecules or graphs or whatever
link |
and find the right one or adjust it to be the right ones.
link |
Sounds like the breast density heuristic from 67,
link |
It's not necessarily that.
link |
It's really, you know, it's really driven by deep understanding.
link |
It's not like they just observe it.
link |
I mean, they do deeply understand chemistry
link |
and they do understand how different groups
link |
and how does it change the properties.
link |
So there is a lot of science that gets into it
link |
and a lot of kind of simulation,
link |
how do you want it to behave?
link |
It's very, very complex.
link |
So they're quite effective at this design, obviously.
link |
Now, effective, yeah, we have drugs.
link |
Like depending on how do you measure effective?
link |
If you measure, it's in terms of cost, it's prohibitive.
link |
If you measure it in terms of times, you know,
link |
we have lots of diseases for which we don't have any drugs
link |
and we don't even know how to approach.
link |
I don't need to mention few drugs
link |
or degenerative disease drugs that fail, you know.
link |
So there are lots of, you know, trials that fail,
link |
you know, in later stages,
link |
which is really catastrophic from the financial perspective.
link |
So, you know, is it the effective,
link |
the most effective mechanism?
link |
Absolutely no, but this is the only one that currently works.
link |
And I would, you know, I was closely interacting
link |
with people in pharmaceutical industry.
link |
I was really fascinating on how sharp
link |
and what a deep understanding of the domain do they have.
link |
It's not observation driven.
link |
There is really a lot of science behind what they do.
link |
But if you ask me, can machine learning change it?
link |
I firmly believe yes,
link |
because even the most experienced chemists cannot, you know,
link |
hold in their memory and understanding
link |
everything that you can learn, you know,
link |
from millions of molecules and reactions.
link |
And the space of graphs is a totally new space.
link |
I mean, it's a really interesting space
link |
for machine learning to explore, graph generation.
link |
Yeah, so there are a lot of things that you can do here.
link |
So we do a lot of work.
link |
So the first tool that we started with
link |
was the tool that can predict properties of the molecules.
link |
So you can just give the molecule and the property.
link |
It can be bioactivity properties.
link |
Or it can be some other property.
link |
And you train the molecules and you can now take a new molecule
link |
and predict this property.
link |
Now, when people started working in this area,
link |
it is something very simple.
link |
They do kind of existing, you know, fingerprints,
link |
which is kind of handcrafted features of the molecule
link |
when you break the graph to substructures
link |
and then you run, I don't know, a feedforward neural network.
link |
And what was interesting to see that clearly, you know,
link |
this was not the most effective way to proceed.
link |
And you need to have much more complex models
link |
that can induce a representation
link |
which can translate this graph into the embeddings
link |
and do these predictions.
link |
So this is one direction.
link |
And another direction, which is kind of related,
link |
is not only to stop by looking at the embedding itself,
link |
but actually modify it to produce better molecules.
link |
So you can think about it as the machine translation
link |
that you can start with a molecule
link |
and then there is an improved version of molecule.
link |
And you can again, with encoder,
link |
translate it into the hidden space
link |
and then learn how to modify it to improve
link |
the in some ways version of the molecules.
link |
So that's, it's kind of really exciting.
link |
We already have seen that the property prediction works
link |
pretty well and now we are generating molecules
link |
and there is actually labs
link |
which are manufacturing this molecule.
link |
So we'll see why it will get to us.
link |
Okay, that's really exciting.
link |
There's a lot of problems.
link |
Speaking of machine translation and embeddings,
link |
you have done a lot of really great research in NLP,
link |
natural language processing.
link |
Can you tell me your journey through NLP,
link |
what ideas, problems, approaches were you working on
link |
were you fascinated with, did you explore
link |
before this magic of deep learning reemerged and after?
link |
So when I started my work in NLP, it was in 97.
link |
This was a very interesting time.
link |
It was exactly the time that I came to ACL
link |
and the dynamic would barely understand English.
link |
But it was exactly like the transition point
link |
because half of the papers were really rule based approaches
link |
where people took more kind of heavy linguistic approaches
link |
for small domains and try to build up from there.
link |
And then there were the first generation of papers
link |
which were corpus based papers.
link |
And they were very simple in our terms
link |
when you collect some statistics
link |
and do prediction based on them.
link |
But I found it really fascinating that one community
link |
can think so very differently about the problem.
link |
And I remember my first papers that I wrote,
link |
it didn't have a single formula,
link |
it didn't have evaluation, it just had examples of outputs.
link |
And this was a standard of the first generation
link |
of the field at a time.
link |
In some ways, I mean, people maybe just started emphasizing
link |
the empirical evaluation,
link |
but for many applications like summarization,
link |
you just wrote some examples of outputs.
link |
And then increasingly you can see
link |
that how the statistical approach has dominated the field.
link |
And we've seen increased performance
link |
across many basic tasks.
link |
The sad part of the story may be that if you look
link |
again through this journey,
link |
we see that the role of linguistics
link |
in some ways greatly diminishes.
link |
And I think that you really need to look
link |
through the whole proceeding to find one or two papers
link |
which make some interesting linguistic references.
link |
This was definitely...
link |
Things like syntactic trees,
link |
just even basically against our conversation
link |
about human understanding of language,
link |
which I guess what linguistics would be,
link |
structured hierarchical representing language
link |
in a way that's human explainable,
link |
understandable is missing today.
link |
I don't know if it is,
link |
what is explainable and understandable.
link |
At the end, we perform functions
link |
and it's okay to have machine which performs a function.
link |
Like when you're thinking about your calculator, correct?
link |
Your calculator can do calculation
link |
very different from you would do the calculation,
link |
but it's very effective in it.
link |
If we can achieve certain tasks with high accuracy,
link |
it doesn't necessarily mean that it has to understand
link |
in the same way as we understand.
link |
In some ways, it's even naive to request
link |
because you have so many other sources of information
link |
that are absent when you are training your system.
link |
And I would tell you one application
link |
that's just really fascinating.
link |
In 97, when it came to ACL,
link |
there were some papers on machine translation.
link |
They were like primitive,
link |
like people were trying really, really simple.
link |
And the feeling, my feeling was that,
link |
to make real machine translation system,
link |
it's like to fly at the moon and build a house there
link |
and the garden and live happily ever after.
link |
I mean, it's like impossible.
link |
I never could imagine that within 10 years,
link |
we would already see the system working.
link |
And now nobody is even surprised
link |
to utilize the system on daily basis.
link |
So this was like a huge, huge progress,
link |
saying that people for very long time
link |
tried to solve using other mechanisms
link |
and they were unable to solve it.
link |
That's why I'm coming back to a question about biology,
link |
that in linguistics, people try to go this way
link |
and try to write the syntactic trees
link |
and try to obstruct it
link |
and to find the right representation.
link |
And, you know, they couldn't get very far
link |
with this understanding while these models,
link |
using, you know, other sources actually capable
link |
to make a lot of progress.
link |
Now, I'm not naive to think
link |
that we are in this paradise space in NLP
link |
and I'm sure as you know,
link |
that when we slightly change the domain
link |
and when we decrease the amount of training,
link |
it can do like really bizarre and funny thing.
link |
But I think it's just a matter of improving
link |
generalization capacity,
link |
which is just a technical question.
link |
Well, so that's the question.
link |
How much of language understanding
link |
can be solved with deep neural networks?
link |
In your intuition, I mean, it's unknown, I suppose.
link |
But as we start to creep towards romantic notions
link |
of the spirit of the Turing test
link |
and conversation and dialogue
link |
and something that may be to me or to us,
link |
so the humans feels like it needs real understanding.
link |
How much can I be achieved
link |
with these neural networks or statistical methods?
link |
So I guess I am very much driven by the outcomes.
link |
Can we achieve the performance,
link |
which would be satisfactory for us for different tasks.
link |
Now, if you again look at machine translation system,
link |
which are trained on large amounts of data,
link |
they really can do a remarkable job
link |
relatively to where they've been a few years ago.
link |
And if you project into the future,
link |
if it will be the same speed of improvement,
link |
Now, does it bother me
link |
that it's not doing the same translation as we are doing?
link |
Now, if you go to cognitive science,
link |
we still don't really understand what we are doing.
link |
I mean, there are a lot of theories
link |
and there is obviously a lot of progress and studying,
link |
but our understanding what exactly goes on in our brains
link |
when we process language is still not crystal clear
link |
and precise that we can translate it into machines.
link |
What does bother me is that, again,
link |
that machines can be extremely brittle
link |
when you go out of your comfort zone of there,
link |
when there is a distributional shift
link |
between training and testing.
link |
And it have been years and years,
link |
every year when they teach a NLP class,
link |
show them some examples of translation
link |
from some newspaper in Hebrew,
link |
whatever, it was perfect.
link |
And then they have a recipe
link |
that Tomi Akala's system sent me a while ago
link |
and it was written in Finnish of Carillian pies.
link |
And it's just a terrible translation.
link |
You cannot understand anything what it does.
link |
It's not like some syntactic mistakes.
link |
It's just terrible.
link |
And year after year, I tried it and will translate it.
link |
And year after year, it does this terrible work
link |
because I guess the recipes are not big part
link |
of their training repertoire.
link |
So, but in terms of outcomes,
link |
that's a really clean, good way to look at it.
link |
I guess the question I was asking is,
link |
do you think, imagine a future,
link |
do you think the current approaches
link |
can pass the Turing test in the way,
link |
in the best possible formulation of the Turing test?
link |
Which is, would you want to have a conversation
link |
with a neural network for an hour?
link |
Oh God, no, no, there are not that many people
link |
that I would want to talk for an hour.
link |
But there are some people in this world, alive or not,
link |
that you would like to talk to for an hour,
link |
could a neural network achieve that outcome?
link |
So I think it would be really hard
link |
to create a successful training set,
link |
which would enable it to have a conversation
link |
for an intercontextual conversation for an hour.
link |
So you think it's a problem of data, perhaps?
link |
I think in some ways it's an important data.
link |
It's a problem both of data and the problem
link |
of the way we are training our systems,
link |
their ability to truly to generalize,
link |
to be very compositional, in some ways, it's limited,
link |
in the current capacity, at least.
link |
You know, we can translate well,
link |
we can find information well, we can extract information.
link |
So there are many capacities in which it's doing very well.
link |
And you can ask me, would you trust the machine
link |
to translate for you and use it as a source?
link |
I would say absolutely, especially if we're talking
link |
about newspaper data or other data,
link |
which is in the realm of its own training set,
link |
But, you know, having conversations with the machine,
link |
it's not something that I would choose to do.
link |
But you know, I would tell you something,
link |
talking about Turing tests
link |
and about all this kind of ELISA conversations.
link |
I remember visiting Tencent in China
link |
and they have this chat board.
link |
And they claim that it is like really humongous amount
link |
of the local population,
link |
which like for hours talks to the chat board,
link |
to me it was, I cannot believe it,
link |
but apparently it's like documented
link |
that there are some people who enjoy this conversation.
link |
And you know, it brought to me another MIT story
link |
about ELISA and Weizimbau.
link |
I don't know if you're familiar with the story.
link |
So Weizimbau was a professor at MIT
link |
and when he developed this ELISA,
link |
which was just doing string matching, very trivial,
link |
like restating of what you said,
link |
with very few rules, no syntax.
link |
Apparently there were secretaries at MIT
link |
that would sit for hours and converse with this trivial thing.
link |
And at the time there was no beautiful interfaces.
link |
So you actually need to go through the pain of communicating.
link |
And Weizimbau himself was so horrified by this phenomenon
link |
that people can believe enough to the machine
link |
that you just need to give them the hint
link |
that machine understands you
link |
and you can complete the rest.
link |
So he kind of stopped this research
link |
and went into kind of trying to understand
link |
what this artificial intelligence can do to our brains.
link |
So my point is, you know, how much,
link |
it's not how good is the technology,
link |
it's how ready we are to believe
link |
that it delivers the good that we are trying to get.
link |
That's a really beautiful way to put it.
link |
I, by the way, I'm not horrified by that possibility
link |
but inspired by it because, I mean, human connection,
link |
whether it's through language or through love,
link |
it seems like it's very amenable to machine learning
link |
and the rest is just challenges of psychology.
link |
Like you said, the secretaries who enjoy spending hours,
link |
I would say I would describe most of our lives
link |
as enjoying spending hours with those we love
link |
for very silly reasons.
link |
All we're doing is keyword matching as well.
link |
So I'm not sure how much intelligence
link |
we exhibit to each other with the people we love
link |
that we're close with.
link |
So it's a very interesting point
link |
of what it means to pass the Turing test with language.
link |
I think you're right.
link |
In terms of conversation,
link |
I think machine translation has very clear performance
link |
and improvement, right?
link |
What it means to have a fulfilling conversation
link |
is very, very person dependent
link |
and context dependent and so on.
link |
That's, yeah, it's very well put.
link |
So, but in your view,
link |
what's a benchmark in natural language, a test,
link |
that's just out of reach right now,
link |
but we might be able to, that's exciting.
link |
Is it in machine, isn't perfecting machine translation
link |
or is there other, is it summarization?
link |
What's out there just out of reach?
link |
It goes across specific application.
link |
It's more about the ability to learn
link |
from few examples for real,
link |
what we call future planning and all these cases.
link |
Because, you know, the way we publish these papers today,
link |
we say, if we have like naively, we get 55,
link |
but now we had a few example and we can move to 65.
link |
None of these methods actually
link |
realistically doing anything useful.
link |
You cannot use them today.
link |
And the ability to be able to generalize and to move
link |
or to be autonomous in finding the data
link |
that you need to learn,
link |
to be able to perfect new tasks or new language.
link |
This is an area where I think we really need
link |
to move forward to and we are not yet there.
link |
Are you at all excited,
link |
curious by the possibility of creating human level intelligence?
link |
Is this, because you've been very in your discussion.
link |
So if we look at oncology,
link |
you're trying to use machine learning to help the world
link |
in terms of alleviating suffering.
link |
If you look at natural language processing,
link |
you're focused on the outcomes of improving practical things
link |
like machine translation.
link |
But, you know, human level intelligence is a thing
link |
that our civilizations dream about creating
link |
super human level intelligence.
link |
Do you think about this?
link |
Do you think it's at all within our reach?
link |
So as you said yourself earlier,
link |
talking about, you know, how do you perceive,
link |
you know, our communications with each other
link |
that, you know, we're matching keywords
link |
and certain behaviors and so on.
link |
So at the end, whenever one assesses,
link |
let's say relations with another person,
link |
you have separate kind of measurements and outcomes
link |
inside your head that determine, you know,
link |
what is the status of the relation.
link |
So one way, this is this classical level.
link |
What is the intelligence?
link |
Is it the fact that now we are going to do
link |
the same way as human is doing
link |
when we don't even understand what the human is doing?
link |
Or we now have an ability to deliver these outcomes,
link |
but not in one area, not in an LPL,
link |
not just to translate or just to answer questions,
link |
but across many, many areas that we can achieve
link |
the functionalities that humans can achieve
link |
with their ability to learn and do other things.
link |
I think this is, and this we can actually measure
link |
how far we are, and that's what makes me excited
link |
that we, you know, in my lifetime,
link |
at least so far what we've seen,
link |
it's like tremendous progress across
link |
with these different functionalities.
link |
And I think it will be really exciting
link |
to see where we will be.
link |
And again, one way to think about is there are machines
link |
which are improving their functionality.
link |
Another one is to think about us with our brains,
link |
which are imperfect, how they can be accelerated
link |
by this technology as it becomes stronger and stronger.
link |
Coming back to another book that I love,
link |
Flowers for Algernon, have you read this book?
link |
You know, there is this point that the patient gets
link |
this miracle cure which changes his brain
link |
and all of a sudden they see life in a different way
link |
and can do certain things better,
link |
but certain things much worse.
link |
So you can imagine this kind of computer augmented cognition
link |
where it can bring you that now in the same way
link |
as, you know, the cars enable us to get to places
link |
where we've never been before.
link |
Can we think differently?
link |
Can we think faster?
link |
So, and we already see a lot of it happening
link |
in how it impacts us,
link |
but I think we have a long way to go there.
link |
So that's sort of artificial intelligence
link |
and technology affecting our,
link |
augmenting our intelligence as humans.
link |
Yesterday, a company called Neuralink announced
link |
they did this whole demonstration.
link |
I don't know if you saw it.
link |
It's, they demonstrated brain, computer,
link |
brain machine interface where there's like a sewing machine
link |
Do you, you know, a lot of that is quite out there
link |
in terms of things that some people would say are impossible,
link |
but they're dreamers and want to engineer systems like that.
link |
Do you see, based on what you just said,
link |
a hope for that more direct interaction with the brain?
link |
I think there are different ways.
link |
One is a direct interaction with the brain.
link |
And again, there are lots of companies
link |
that work in this space.
link |
And I think there will be a lot of developments.
link |
When I'm just thinking that many times
link |
we are not aware of our feelings
link |
of motivation, what drives us.
link |
Like let me give you a trivial example, our attention.
link |
There are a lot of studies that demonstrate
link |
that it takes a while to a person to understand
link |
that they are not attentive anymore.
link |
And we know that there are people
link |
who really have strong capacity to hold attention.
link |
There are another end of the spectrum,
link |
people with ADD and other issues
link |
that they have problem to regulate their attention.
link |
Imagine to yourself that you have like a cognitive aid
link |
that just alerts you based on your gaze.
link |
That your attention is now not on what you are doing.
link |
And instead of writing a paper, you're now dreaming
link |
of what you're gonna do in the evening.
link |
So even this kind of simple measurement things,
link |
how they can change us.
link |
And I see it even in the simple ways with myself.
link |
I have my zone up from that I got in MIT gym.
link |
It kind of records how much did you run
link |
and you have some points and you can get some status,
link |
Like I said, what is this ridiculous thing?
link |
Who would ever care about some status in some arm?
link |
So to maintain the status,
link |
you have to set a number of points every month.
link |
And not only is that they do it every single month
link |
for the last 18 months,
link |
it went to the point that I was injured.
link |
And when I could run again,
link |
I in two days, I did like some humongous amount
link |
of writing just to complete the points.
link |
It was like really not safe.
link |
It's like, I'm not gonna lose my status
link |
because I want to get there.
link |
So you can already see that this direct measurement
link |
and the feedback, we're looking at video games
link |
and see why the addiction aspect of it,
link |
but you can imagine that the same idea
link |
can be expanded to many other areas of our life
link |
when we really can get feedback
link |
and imagine in your case in relations
link |
when we are doing keyword matching,
link |
imagine that the person who is generating the key ones,
link |
that person gets direct feedback
link |
before the whole thing explodes.
link |
Is it maybe at this happy point,
link |
we are going in the wrong direction?
link |
Maybe it will be really a behavior modifying moment.
link |
So yeah, it's a relationship management too.
link |
So yeah, that's a fascinating whole area
link |
of psychology actually as well,
link |
of seeing how our behavior has changed
link |
with basically all human relations
link |
now have other non human entities helping us out.
link |
So you've, you teach a large,
link |
a huge machine learning course here at MIT.
link |
I can ask you a million questions,
link |
but you've seen a lot of students.
link |
What ideas do students struggle with the most
link |
as they first enter this world of machine learning?
link |
Actually, this year was the first time
link |
I started teaching a small machine learning class
link |
and it came as a result of what I saw
link |
in my big machine learning class that Tommy Ackle
link |
and I built maybe six years ago.
link |
What we've seen that as this area become more and more popular,
link |
more and more people at MIT want to take this class.
link |
And while we designed it for computer science majors,
link |
there were a lot of people who really are interested
link |
to learn it, but unfortunately,
link |
their background was not enabling them
link |
to do well in the class.
link |
And many of them associated machine learning
link |
with a world struggle and failure,
link |
primarily for non majors.
link |
And that's why we actually started a new class
link |
which we call machine learning from algorithms to modeling,
link |
which emphasizes more the modeling aspects of it
link |
and focuses on, it has majors and non majors.
link |
So we kind of try to extract the relevant parts
link |
and make it more accessible
link |
because the fact that we're teaching 20 classifiers
link |
in standard machine learning class
link |
is really a big question we really needed.
link |
But it was interesting to see this
link |
from first generation of students,
link |
when they came back from their internships
link |
and from their jobs,
link |
what different and exciting things they can do
link |
is that they would never think
link |
that you can even apply machine learning to.
link |
Some of them are like matching their relations
link |
and other things like variety of different applications.
link |
Everything is amenable to machine learning.
link |
That actually brings up an interesting point
link |
of computer science in general.
link |
It almost seems, maybe I'm crazy,
link |
but it almost seems like everybody needs to learn
link |
how to program these days.
link |
If you're 20 years old or if you're starting school,
link |
even if you're an English major,
link |
it seems like programming
link |
unlocks so much possibility in this world.
link |
So when you interact with those non majors,
link |
is there skills that they were simply lacking at the time
link |
that you wish they had
link |
and that they learned in high school and so on?
link |
Like how should education change
link |
in this computerized world that we live in?
link |
So seeing because they knew that there is a Python component
link |
their Python skills were okay
link |
and the class is not really heavy on programming.
link |
They primarily kind of add parts to the programs.
link |
I think it was more of their mathematical barriers
link |
and the class, again, with the design on the majors
link |
was using the notation like big O for complexity
link |
and others, people who come from different backgrounds
link |
just don't have it in the lexical.
link |
So necessarily very challenging notion,
link |
but they were just not aware.
link |
So I think that, you know, kind of linear algebra
link |
and probability, the basics, the calculus,
link |
want to vary the calculus, things that can help.
link |
What advice would you give to students
link |
interested in machine learning, interested,
link |
if you've talked about detecting curing cancer,
link |
drug design, if they want to get into that field,
link |
what should they do?
link |
Get into it and succeed as researchers
link |
and entrepreneurs.
link |
The first good piece of news is that right now
link |
there are lots of resources
link |
that are created at different levels
link |
and you can find online
link |
on your school classes,
link |
which are more mathematical or more applied and so on.
link |
So you can find a kind of a preacher
link |
which preaches your own language
link |
where you can enter the field
link |
and you can make many different types of contribution
link |
depending of what is your strengths.
link |
And the second point,
link |
I think it's really important to find some area
link |
which you really care about
link |
and it can motivate your learning
link |
and it can be for somebody curing cancer
link |
or doing cell driving cars or whatever,
link |
but to find an area where there is data
link |
where you believe there are strong patterns
link |
and we should be doing it
link |
and we're still not doing it
link |
or you can do it better
link |
and just start there
link |
and see a way it can bring you.
link |
So you've been very successful
link |
in many directions in life,
link |
but you also mentioned Flowers of Argonaut.
link |
And I think I've read or listened to you mention somewhere
link |
that researchers often get lost
link |
in the details of their work.
link |
This is per our original discussion with cancer and so on
link |
and don't look at the bigger picture,
link |
the bigger questions of meaning and so on.
link |
So let me ask you the impossible question
link |
of what's the meaning of this thing,
link |
of life, of your life, of research.
link |
Why do you think we descendant of great apes
link |
are here on this spinning ball?
link |
You know, I don't think that I have really a global answer
link |
you know, maybe that's why I didn't go to humanities
link |
and I didn't take humanities classes in my undergrad.
link |
But the way I am thinking about it,
link |
each one of us inside of them have their own set of,
link |
you know, things that we believe are important.
link |
And it just happens that we are busy
link |
with achieving various goals,
link |
busy listening to others
link |
and to kind of try to conform
link |
to be part of the crowd that we don't listen to that part.
link |
And, you know, we all should find some time to understand
link |
what is our own individual missions
link |
and we may have very different missions
link |
and to make sure that while we are running 10,000 things,
link |
we are not, you know, missing out
link |
and we're putting all the resources
link |
to satisfy our own mission.
link |
And if I look over my time,
link |
when I was younger, most of these missions,
link |
you know, I was primarily driven by the external stimulus,
link |
you know, to achieve this or to be that.
link |
And now a lot of what I do is driven by really thinking
link |
what is important for me to achieve independently
link |
of the external recognition.
link |
And, you know, I don't mind to be viewed in certain ways.
link |
The most important thing for me is to be true to myself,
link |
to what I think is right.
link |
How long did it take?
link |
How hard was it to find the you that you have to be true to?
link |
So it takes time and even now sometimes, you know,
link |
the vanity and the triviality can take, you know.
link |
Yeah, it can everywhere, you know, it's just the vanity.
link |
The vanity is different, the vanity in different places,
link |
but we all have our piece of vanity.
link |
But I think actually for me,
link |
the many times the place to get back to it is, you know,
link |
when I'm alone and also when I read.
link |
And I think by selecting the right books,
link |
you can get the right questions and learn from what you read.
link |
So, but again, it's not perfect,
link |
like vanity sometimes dominates.
link |
Well, that's a beautiful way to end.
link |
Thank you so much for talking today.
link |
Thank you. That was fun.