Press "Enter" to skip to content

How computers learn like people | Doina Precup | TEDxMcGill


the great magic to be here and telling

the great magic to be here and telling you a little bit about what they do so

one story is that I grew up in Romania

one story is that I grew up in Romania back in the days when it was a communist

back in the days when it was a communist

back in the days when it was a communist country so things were pretty dreary and

country so things were pretty dreary and

country so things were pretty dreary and as a result I know being rebellious

as a result I know being rebellious

as a result I know being rebellious teenagers we did very radical things

teenagers we did very radical things

teenagers we did very radical things like watching pirated sci-fi movies it’s

like watching pirated sci-fi movies it’s

like watching pirated sci-fi movies it’s also very much else radical to do and so

also very much else radical to do and so

also very much else radical to do and so I’m not looking at androids like beta

I’m not looking at androids like beta

I’m not looking at androids like beta and robots and I thought you know this

and robots and I thought you know this

and robots and I thought you know this is really cool I really like about

is really cool I really like about

is really cool I really like about something like that

something like that

something like that so it really inspired me to actually go

so it really inspired me to actually go

so it really inspired me to actually go inside the computer science of the

inside the computer science of the

inside the computer science of the University and one of the first sort of

University and one of the first sort of

University and one of the first sort of significant assignments that I had there

significant assignments that I had there

significant assignments that I had there was to build a program that could play

was to build a program that could play

was to build a program that could play chess now just as a game that people

chess now just as a game that people

chess now just as a game that people consider hallmark of intelligence I

consider hallmark of intelligence I

consider hallmark of intelligence I thought okay this is pretty good we’re

thought okay this is pretty good we’re

thought okay this is pretty good we’re gonna actually make some set to build

gonna actually make some set to build

gonna actually make some set to build some robots another that because my

some robots another that because my

some robots another that because my assignment and these people said well

assignment and these people said well

assignment and these people said well you know think about how do you play

you know think about how do you play

you know think about how do you play chess and then make those rules and

chess and then make those rules and

chess and then make those rules and programs I know did some I’m really a

programs I know did some I’m really a

programs I know did some I’m really a lousy chess player so all I could do is

lousy chess player so all I could do is

lousy chess player so all I could do is basically tell the program some very

basically tell the program some very

basically tell the program some very very basic things about how the pieces

very basic things about how the pieces

very basic things about how the pieces move it wasn’t really very good now I

move it wasn’t really very good now I

move it wasn’t really very good now I decided to not give up and so I went to

decided to not give up and so I went to

decided to not give up and so I went to grad school and at the beginning of my

grad school and at the beginning of my

grad school and at the beginning of my grad school actually a watershed moment

grad school actually a watershed moment

grad school actually a watershed moment for artificial intelligence happened

for artificial intelligence happened

for artificial intelligence happened this is a picture that was in the news

this is a picture that was in the news

this is a picture that was in the news at the time this is Gary Kasparov who

at the time this is Gary Kasparov who

at the time this is Gary Kasparov who was the world’s chess champion beaten by

was the world’s chess champion beaten by

was the world’s chess champion beaten by a computer and the computer was blue was

a computer and the computer was blue was

a computer and the computer was blue was a creation of IBM and really all of us

a creation of IBM and really all of us

a creation of IBM and really all of us were studying AI and we’re very inspired

were studying AI and we’re very inspired

were studying AI and we’re very inspired by this we thought okay now this is true

by this we thought okay now this is true

by this we thought okay now this is true intelligence computer it’s better than a

intelligence computer it’s better than a

intelligence computer it’s better than a person again

person again

person again really consider to be very significant

really consider to be very significant

really consider to be very significant and so no very enthusiastically we

and so no very enthusiastically we

and so no very enthusiastically we welcome to read the papers that describe

welcome to read the papers that describe

welcome to read the papers that describe the technology and then he realized that

the technology and then he realized that

the technology and then he realized that actually this was an amazing feat but it

actually this was an amazing feat but it

actually this was an amazing feat but it was essentially a human feet there was a

was essentially a human feet there was a

was essentially a human feet there was a computer that was purpose-built to play

computer that was purpose-built to play

computer that was purpose-built to play chess that’s all it could do it was

chess that’s all it could do it was

chess that’s all it could do it was really really didn’t playing chess but

really really didn’t playing chess but

really really didn’t playing chess but it was very special and couldn’t do much

it was very special and couldn’t do much

it was very special and couldn’t do much of anything else

of anything else

of anything else also people sort of thought very

also people sort of thought very

also people sort of thought very carefully about how masters played the

carefully about how masters played the

carefully about how masters played the game and they built this capability into

game and they built this capability into

game and they built this capability into the computer so it was amazing but it

the computer so it was amazing but it

the computer so it was amazing but it wasn’t quite what I was looking for now

wasn’t quite what I was looking for now

wasn’t quite what I was looking for now yeah we are 20 years later and actually

yeah we are 20 years later and actually

yeah we are 20 years later and actually things have moved quite a bit so these

things have moved quite a bit so these

things have moved quite a bit so these are some pictures taken from the games

are some pictures taken from the games

are some pictures taken from the games of Lisa goal and kg who played also

of Lisa goal and kg who played also

of Lisa goal and kg who played also computer program this time at the game

computer program this time at the game

computer program this time at the game of God now go is an ancient game that is

of God now go is an ancient game that is

of God now go is an ancient game that is considered significantly harder than

considered significantly harder than

considered significantly harder than chess if they’re all strategy as well as

chess if they’re all strategy as well as

chess if they’re all strategy as well as tactics and it is sort of considerably

tactics and it is sort of considerably

tactics and it is sort of considerably more difficult also because there’s many

more difficult also because there’s many

more difficult also because there’s many many many possible moves and people have

many many possible moves and people have

many many possible moves and people have to think that very close okay capture

to think that very close okay capture

to think that very close okay capture territory how do I defend and so on and

territory how do I defend and so on and

territory how do I defend and so on and it turns out that this program called

it turns out that this program called

it turns out that this program called alpha del and it’s sort of next version

alpha del and it’s sort of next version

alpha del and it’s sort of next version called alpha zero actually was really

called alpha zero actually was really

called alpha zero actually was really really good to the point that it is now

really good to the point that it is now

really good to the point that it is now essentially the world’s best goalie and

essentially the world’s best goalie and

essentially the world’s best goalie and it made the cover of nature and a lot of

it made the cover of nature and a lot of

it made the cover of nature and a lot of us got very excited about it and we now

us got very excited about it and we now

us got very excited about it and we now have in fact a much more interesting way

have in fact a much more interesting way

have in fact a much more interesting way of doing this kind of a feat which is

of doing this kind of a feat which is

of doing this kind of a feat which is not so much engineering so this is what

not so much engineering so this is what

not so much engineering so this is what I want to tell you about how do these

I want to tell you about how do these

I want to tell you about how do these programs actually play the game it turns

programs actually play the game it turns

programs actually play the game it turns out that they’re not so much programs as

out that they’re not so much programs as

out that they’re not so much programs as they’d learn how to do it now when you

they’d learn how to do it now when you

they’d learn how to do it now when you think about learning kids learning

think about learning kids learning

think about learning kids learning comes to mind is going to school okay so

comes to mind is going to school okay so

comes to mind is going to school okay so go to school you have a teacher then eat

go to school you have a teacher then eat

go to school you have a teacher then eat your selfie various things you kind of

your selfie various things you kind of

your selfie various things you kind of ingest on the menu learn how to do stuff

ingest on the menu learn how to do stuff

ingest on the menu learn how to do stuff that’s one way to do learn in in

that’s one way to do learn in in

that’s one way to do learn in in computers it’s called supervised

computers it’s called supervised

computers it’s called supervised learning so you can have for example a

learning so you can have for example a

learning so you can have for example a computer that wants to learn how to

computer that wants to learn how to

computer that wants to learn how to label images how do we do this what will

label images how do we do this what will

label images how do we do this what will show the computer a lot of images where

show the computer a lot of images where

show the computer a lot of images where somebody has gone in and actually

somebody has gone in and actually

somebody has gone in and actually labeled the things that are interesting

labeled the things that are interesting

labeled the things that are interesting so in this case we’re on the computer to

so in this case we’re on the computer to

so in this case we’re on the computer to learn how to label faces and so

learn how to label faces and so

learn how to label faces and so somebody’s going via mentioned drawn

somebody’s going via mentioned drawn

somebody’s going via mentioned drawn some little squares around the faces and

some little squares around the faces and

some little squares around the faces and I’m going to show this information to

I’m going to show this information to

I’m going to show this information to the computer the image is just a set of

the computer the image is just a set of

the computer the image is just a set of pixels and we know what we’re looking

pixels and we know what we’re looking

pixels and we know what we’re looking for in the image and so now we can show

for in the image and so now we can show

for in the image and so now we can show the computer lots and lots and lots of

the computer lots and lots and lots of

the computer lots and lots and lots of images millions of images as it turns

images millions of images as it turns

images millions of images as it turns out and to learn how to recognize faces

out and to learn how to recognize faces

out and to learn how to recognize faces now what does the computer have inside

now what does the computer have inside

now what does the computer have inside it has a little brain for the artificial

it has a little brain for the artificial

it has a little brain for the artificial neural network or a big network so in

neural network or a big network so in

neural network or a big network so in the picture over there it really has

the picture over there it really has

the picture over there it really has some very small units that can do very

some very small units that can do very

some very small units that can do very simple things okay like compute a

simple things okay like compute a

simple things okay like compute a certain line function or maybe like a

certain line function or maybe like a

certain line function or maybe like a little squiggly line function so each of

little squiggly line function so each of

little squiggly line function so each of the circuit is very very simple kind of

the circuit is very very simple kind of

the circuit is very very simple kind of like the human neuron the power comes

like the human neuron the power comes

like the human neuron the power comes from the connections the connections

from the connections the connections

from the connections the connections have numbers on them and these numbers

have numbers on them and these numbers

have numbers on them and these numbers get trained some of them get Kai out

get trained some of them get Kai out

get trained some of them get Kai out some of them get low and each of these

some of them get low and each of these

some of them get low and each of these little units that’s a very specialized

little units that’s a very specialized

little units that’s a very specialized kind of thing so for example some of

kind of thing so for example some of

kind of thing so for example some of them are going to just simply learn to

them are going to just simply learn to

them are going to just simply learn to recognize edges or lines

recognize edges or lines

recognize edges or lines some of them are going to actually

some of them are going to actually

some of them are going to actually recognize certain kinds of faces turns

recognize certain kinds of faces turns

recognize certain kinds of faces turns out some of these names actually learn

out some of these names actually learn

out some of these names actually learn how to recognize cats okay why well

how to recognize cats okay why well

how to recognize cats okay why well because a lot of damages on the internet

because a lot of damages on the internet

because a lot of damages on the internet are cats and dogs and so on and so we

are cats and dogs and so on and so we

are cats and dogs and so on and so we now have these wonderful artificial

now have these wonderful artificial

now have these wonderful artificial brains that are really really good at

brains that are really really good at

brains that are really really good at recognizing cats from all angles and all

recognizing cats from all angles and all

recognizing cats from all angles and all different breeds so this is interesting

different breeds so this is interesting

different breeds so this is interesting this is part of the way how alpha bill

this is part of the way how alpha bill

this is part of the way how alpha bill does things in fact okay so what I mean

does things in fact okay so what I mean

does things in fact okay so what I mean by that initially after though estranged

by that initially after though estranged

by that initially after though estranged by looking at a lot of games that people

by looking at a lot of games that people

by looking at a lot of games that people play good Grandmaster level players and

play good Grandmaster level players and

play good Grandmaster level players and essentially thinking the move that they

essentially thinking the move that they

essentially thinking the move that they did was the right to move okay

did was the right to move okay

did was the right to move okay and so the program learns how to refuse

and so the program learns how to refuse

and so the program learns how to refuse it use these moves now this is great

it use these moves now this is great

it use these moves now this is great it’s very interesting it’s a very

it’s very interesting it’s a very

it’s very interesting it’s a very powerful program but there’s one catch

powerful program but there’s one catch

powerful program but there’s one catch here which is this requires a lot of

here which is this requires a lot of

here which is this requires a lot of data that is provided by people okay

data that is provided by people okay

data that is provided by people okay with the images we need millions of

with the images we need millions of

with the images we need millions of images labeled by people now people get

images labeled by people now people get

images labeled by people now people get bored very quickly of this kind of task

bored very quickly of this kind of task

bored very quickly of this kind of task and then subjects is like a mechanical

and then subjects is like a mechanical

and then subjects is like a mechanical there actually aren’t that many games

there actually aren’t that many games

there actually aren’t that many games that are played by grandmasters that we

that are played by grandmasters that we

that are played by grandmasters that we can look at for information so now if we

can look at for information so now if we

can look at for information so now if we think about how people learn yes you

think about how people learn yes you

think about how people learn yes you learn things in school but actually

learn things in school but actually

learn things in school but actually you’re going types of stuff even before

you’re going types of stuff even before

you’re going types of stuff even before you go yourself you learn how to

you go yourself you learn how to

you go yourself you learn how to recognize objects and what colors are

recognize objects and what colors are

recognize objects and what colors are you don’t have to count and you learn to

you don’t have to count and you learn to

you don’t have to count and you learn to interact through the world in various

interact through the world in various

interact through the world in various ways and so that’s the kind of process

ways and so that’s the kind of process

ways and so that’s the kind of process we would like to mimic in a computer

we would like to mimic in a computer

we would like to mimic in a computer it’s a different kind of learning called

it’s a different kind of learning called

it’s a different kind of learning called reinforcement learning and it turns out

reinforcement learning and it turns out

reinforcement learning and it turns out that this is not even just something

that this is not even just something

that this is not even just something that people do but animals do it it’s

that people do but animals do it it’s

that people do but animals do it it’s very widespread in the natural world and

very widespread in the natural world and

very widespread in the natural world and so in one picture there you have a build

so in one picture there you have a build

so in one picture there you have a build Mouse

Mouse

Mouse that’s learning a task the task is

that’s learning a task the task is

that’s learning a task the task is essentially cupressus sequence of levers

essentially cupressus sequence of levers

essentially cupressus sequence of levers in order to get some coolant and the

in order to get some coolant and the

in order to get some coolant and the mouth just kind of dust things up random

mouth just kind of dust things up random

mouth just kind of dust things up random stumbles upon believers when it gets the

stumbles upon believers when it gets the

stumbles upon believers when it gets the right femur as it gets the food and

right femur as it gets the food and

right femur as it gets the food and that’s its reward okay that feels good

that’s its reward okay that feels good

that’s its reward okay that feels good so no smells it’s happy and after what

so no smells it’s happy and after what

so no smells it’s happy and after what kind of triggers a path of learning that

kind of triggers a path of learning that

kind of triggers a path of learning that makes it remember what were these things

makes it remember what were these things

makes it remember what were these things I get that like me today reward

I get that like me today reward

I get that like me today reward it turns out we can train computers in

it turns out we can train computers in

it turns out we can train computers in much the same way through reinforcement

much the same way through reinforcement

much the same way through reinforcement for it so you think of the computer

for it so you think of the computer

for it so you think of the computer program now as being exposed to the

program now as being exposed to the

program now as being exposed to the world it watches the world it can do

world it watches the world it can do

world it watches the world it can do things it’s an outreach servant and as a

things it’s an outreach servant and as a

things it’s an outreach servant and as a result of what it does it may receive a

result of what it does it may receive a

result of what it does it may receive a reward now this could be a positive

reward now this could be a positive

reward now this could be a positive reward number and number and the whole

reward number and number and the whole

reward number and number and the whole the computer is to find actions that

the computer is to find actions that

the computer is to find actions that essentially give it as large a number as

essentially give it as large a number as

essentially give it as large a number as possible

possible

possible that’s a bias that we put it but it’s

that’s a bias that we put it but it’s

that’s a bias that we put it but it’s never told the actual thing that it

never told the actual thing that it

never told the actual thing that it should do it’s allowed to experiment

should do it’s allowed to experiment

should do it’s allowed to experiment the other interesting thing is that

the other interesting thing is that

the other interesting thing is that these numbers might come much later than

these numbers might come much later than

these numbers might come much later than why the action has happened for those of

why the action has happened for those of

why the action has happened for those of you who are doing the University you

you who are doing the University you

you who are doing the University you probably you know expect some reward at

probably you know expect some reward at

probably you know expect some reward at the end of the day when you get your

the end of the day when you get your

the end of the day when you get your degree if it’s going to take some years

degree if it’s going to take some years

degree if it’s going to take some years to get there and so it won’t anybody

to get there and so it won’t anybody

to get there and so it won’t anybody make it somehow they still be enforce

make it somehow they still be enforce

make it somehow they still be enforce our actions so that’s in fact how

our actions so that’s in fact how

our actions so that’s in fact how alphago learns it looks at the board

alphago learns it looks at the board

alphago learns it looks at the board it tries something that’s an actions

it tries something that’s an actions

it tries something that’s an actions there’s no feedback until the end of the

there’s no feedback until the end of the

there’s no feedback until the end of the game yeah

game yeah

game yeah it observes whether it’s one big a

it observes whether it’s one big a

it observes whether it’s one big a monstrosity and if it’s won the game

monstrosity and if it’s won the game

monstrosity and if it’s won the game it’s going to make all of the actions

it’s going to make all of the actions

it’s going to make all of the actions that happen during that game a little

that happen during that game a little

that happen during that game a little bit more likely because they obviously

bit more likely because they obviously

bit more likely because they obviously let’s till the dolls come and if the

let’s till the dolls come and if the

let’s till the dolls come and if the game will stop all those actions are

game will stop all those actions are

game will stop all those actions are going to become a little bit less likely

going to become a little bit less likely

going to become a little bit less likely so that’s the basic principle and not

so that’s the basic principle and not

so that’s the basic principle and not the program is just going to sit there

the program is just going to sit there

the program is just going to sit there in its happy computer corner and just

in its happy computer corner and just

in its happy computer corner and just blame millions and millions of games

blame millions and millions of games

blame millions and millions of games bites up it turns out that this leads

bites up it turns out that this leads

bites up it turns out that this leads the best program in the world not only

the best program in the world not only

the best program in the world not only that but because the program is not

that but because the program is not

that but because the program is not really told what’s the right thing to do

really told what’s the right thing to do

really told what’s the right thing to do because it learns by experimentation it

because it learns by experimentation it

because it learns by experimentation it can in fact invent new ways of playing

can in fact invent new ways of playing

can in fact invent new ways of playing that are different from what people

that are different from what people

that are different from what people believe so this is a graph that’s

believe so this is a graph that’s

believe so this is a graph that’s showing after those year or learning

showing after those year or learning

showing after those year or learning performance is high the higher the

performance is high the higher the

performance is high the higher the better so this is it’s it’s renting and

better so this is it’s it’s renting and

better so this is it’s it’s renting and after three days it actually surpasses

after three days it actually surpasses

after three days it actually surpasses the ability of our goalie this is the

the ability of our goalie this is the

the ability of our goalie this is the version that we lease at all they’ll

version that we lease at all they’ll

version that we lease at all they’ll strength much by supervised learning and

strength much by supervised learning and

strength much by supervised learning and then after 21 days it actually reaches

then after 21 days it actually reaches

then after 21 days it actually reaches the level of master which is the one

the level of master which is the one

the level of master which is the one that defeated key G and at the end of

that defeated key G and at the end of

that defeated key G and at the end of the training after 40 days it becomes so

the training after 40 days it becomes so

the training after 40 days it becomes so good that it beats the previous version

good that it beats the previous version

good that it beats the previous version a hundred games to nothing so it’s

a hundred games to nothing so it’s

a hundred games to nothing so it’s absolutely the best go player in the

absolutely the best go player in the

absolutely the best go player in the world what else is interesting about

world what else is interesting about

world what else is interesting about this program is that now there’s no

this program is that now there’s no

this program is that now there’s no special purpose hardware ok turns on the

special purpose hardware ok turns on the

special purpose hardware ok turns on the regular computer and the algorithm is

regular computer and the algorithm is

regular computer and the algorithm is actually very general there’s nothing in

actually very general there’s nothing in

actually very general there’s nothing in the algorithm that told the computer

the algorithm that told the computer

the algorithm that told the computer that is specific

that is specific

that is specific though we just know that you should

though we just know that you should

though we just know that you should prefer to win and now go up and display

prefer to win and now go up and display

prefer to win and now go up and display out the games and figure it out so

out the games and figure it out so

out the games and figure it out so that’s a very powerful concept because

that’s a very powerful concept because

that’s a very powerful concept because it generalizes to many many different

it generalizes to many many different

it generalizes to many many different situations now

situations now

situations now of course I’m very happy with games and

of course I’m very happy with games and

of course I’m very happy with games and many of us computer scientists are very

many of us computer scientists are very

many of us computer scientists are very happy with games but there’s actually

happy with games but there’s actually

happy with games but there’s actually real applications to this and so in our

real applications to this and so in our

real applications to this and so in our lab we look at all kinds of ways that

lab we look at all kinds of ways that

lab we look at all kinds of ways that artificial intelligence and these kinds

artificial intelligence and these kinds

artificial intelligence and these kinds of learning methods can be applied one

of learning methods can be applied one

of learning methods can be applied one of the its pictured here is the project

of the its pictured here is the project

of the its pictured here is the project we have going on at the Children’s

we have going on at the Children’s

we have going on at the Children’s Hospital which is looking at very

Hospital which is looking at very

Hospital which is looking at very premature babies so in this case the

premature babies so in this case the

premature babies so in this case the babies are born they’re very small they

babies are born they’re very small they

babies are born they’re very small they cannot breathe on their own and so

cannot breathe on their own and so

cannot breathe on their own and so they’re hooked up to a ventilator and

they’re hooked up to a ventilator and

they’re hooked up to a ventilator and now we would like to figure out when

now we would like to figure out when

now we would like to figure out when they are they actually ready to be taken

they are they actually ready to be taken

they are they actually ready to be taken off of that ventilator machine that’s a

off of that ventilator machine that’s a

off of that ventilator machine that’s a decision done by a doctor but the doctor

decision done by a doctor but the doctor

decision done by a doctor but the doctor needs help because actually they’re not

needs help because actually they’re not

needs help because actually they’re not always there to observe the baby and so

always there to observe the baby and so

always there to observe the baby and so they have limited amounts of information

they have limited amounts of information

they have limited amounts of information and also there’s a lot of noise

and also there’s a lot of noise

and also there’s a lot of noise that’s a thankful setting the babies get

that’s a thankful setting the babies get

that’s a thankful setting the babies get picked up and put down the sensors lose

picked up and put down the sensors lose

picked up and put down the sensors lose contact and so on so what we do is we

contact and so on so what we do is we

contact and so on so what we do is we gather data in the hospital setting and

gather data in the hospital setting and

gather data in the hospital setting and then we want a computer to essentially

then we want a computer to essentially

then we want a computer to essentially predict their risk that the baby would

predict their risk that the baby would

predict their risk that the baby would have incurred if it were to be extubated

have incurred if it were to be extubated

have incurred if it were to be extubated right now and in this case the computer

right now and in this case the computer

right now and in this case the computer is not actually experimenting with

is not actually experimenting with

is not actually experimenting with taking actions

taking actions

taking actions it’s simply learning to predict the

it’s simply learning to predict the

it’s simply learning to predict the right outcome and it’s being rewarded

right outcome and it’s being rewarded

right outcome and it’s being rewarded for predicting the right now there’s

for predicting the right now there’s

for predicting the right now there’s another application that I’ll show you

another application that I’ll show you

another application that I’ll show you here which is we have some data image

here which is we have some data image

here which is we have some data image data that’s very corrupted I’m not sure

data that’s very corrupted I’m not sure

data that’s very corrupted I’m not sure if you can tell what’s in this image

if you can tell what’s in this image

if you can tell what’s in this image okay it’s a little bit hard to tell but

okay it’s a little bit hard to tell but

okay it’s a little bit hard to tell but if I let the computer do its job okay

if I let the computer do its job okay

if I let the computer do its job okay you see some digits up here

you see some digits up here

you see some digits up here image now this is a task that’s called

image now this is a task that’s called

image now this is a task that’s called fixing corrupting data our data

fixing corrupting data our data

fixing corrupting data our data imputation in statistical terms it’s a

imputation in statistical terms it’s a

imputation in statistical terms it’s a problem where you have to look at the

problem where you have to look at the

problem where you have to look at the data that’s there in order to kind of

data that’s there in order to kind of

data that’s there in order to kind of get us but what’s not there and people

get us but what’s not there and people

get us but what’s not there and people are good at this putting a little bit of

are good at this putting a little bit of

are good at this putting a little bit of a guys missing but they’re not so good

a guys missing but they’re not so good

a guys missing but they’re not so good at when a lot of data is missing and you

at when a lot of data is missing and you

at when a lot of data is missing and you can kind of see that it’s going to be

can kind of see that it’s going to be

can kind of see that it’s going to be hard for you to recognize these images

hard for you to recognize these images

hard for you to recognize these images have to kind of fill filling the gaps

have to kind of fill filling the gaps

have to kind of fill filling the gaps without any help so what the computer

without any help so what the computer

without any help so what the computer does is it looks at the image it looks

does is it looks at the image it looks

does is it looks at the image it looks at the context of what’s missing and can

at the context of what’s missing and can

at the context of what’s missing and can decide to fill in a pixel or not and it

decide to fill in a pixel or not and it

decide to fill in a pixel or not and it kind of does this over time little by

kind of does this over time little by

kind of does this over time little by little and again it’s rewarded by

little and again it’s rewarded by

little and again it’s rewarded by obtaining at the end the right answer

obtaining at the end the right answer

obtaining at the end the right answer but once it’s trained it can do this on

but once it’s trained it can do this on

but once it’s trained it can do this on any other kinds of images so in the

any other kinds of images so in the

any other kinds of images so in the context of this presentation of giving

context of this presentation of giving

context of this presentation of giving voice I think it’s really important for

voice I think it’s really important for

voice I think it’s really important for us to think of how we’re going to use

us to think of how we’re going to use

us to think of how we’re going to use these very powerful tools now in our

these very powerful tools now in our

these very powerful tools now in our lives they are available some of you

lives they are available some of you

lives they are available some of you probably use them every day if you

probably use them every day if you

probably use them every day if you interact with with your phone you talk

interact with with your phone you talk

interact with with your phone you talk to your phone your phone that some voice

to your phone your phone that some voice

to your phone your phone that some voice recognition if you do translation google

recognition if you do translation google

recognition if you do translation google translate you’re going to interact with

translate you’re going to interact with

translate you’re going to interact with these kinds of methods how are you going

these kinds of methods how are you going

these kinds of methods how are you going to make use of this so that this

to make use of this so that this

to make use of this so that this technology can actually help us improve

technology can actually help us improve

technology can actually help us improve and I hope that all of you can think of

and I hope that all of you can think of

and I hope that all of you can think of some useful ways it

some useful ways it

some useful ways it [Applause]

”

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *