so you’ve just been told that I’m a lecturer and as such I get to teach but
lecturer and as such I get to teach but
lecturer and as such I get to teach but I also get to have really interesting
I also get to have really interesting
I also get to have really interesting meetings with very bright students about
meetings with very bright students about
meetings with very bright students about smaller and bigger projects and this one
smaller and bigger projects and this one
smaller and bigger projects and this one is one of those just from about a week
is one of those just from about a week
is one of those just from about a week ago now we were discussing some project
ago now we were discussing some project
ago now we were discussing some project that I would actually tell you about and
that I would actually tell you about and
that I would actually tell you about and if you look at it you will see some
if you look at it you will see some
if you look at it you will see some bright happy students and you will also
bright happy students and you will also
bright happy students and you will also see the standard laptops on on the table
see the standard laptops on on the table
see the standard laptops on on the table but you also see something which is a
but you also see something which is a
but you also see something which is a little unusual and if i zoom in you will
little unusual and if i zoom in you will
little unusual and if i zoom in you will see better now this just make sure we
see better now this just make sure we
see better now this just make sure we understand each other that this is not
understand each other that this is not
understand each other that this is not how I normally have my student meetings
how I normally have my student meetings
how I normally have my student meetings but this was actually brought by the
but this was actually brought by the
but this was actually brought by the students to celebrate a special occasion
students to celebrate a special occasion
students to celebrate a special occasion I knew nothing about when when I arrived
I knew nothing about when when I arrived
I knew nothing about when when I arrived when I entered the room and I was
when I entered the room and I was
when I entered the room and I was surprised I was actually very pleasantly
surprised I was actually very pleasantly
surprised I was actually very pleasantly surprised and I was surprised because
surprised and I was surprised because
surprised and I was surprised because this is not normal yeah this is unusual
this is not normal yeah this is unusual
this is not normal yeah this is unusual I don’t normally enter a meeting with my
I don’t normally enter a meeting with my
I don’t normally enter a meeting with my students and a bottle of wine and is
students and a bottle of wine and is
students and a bottle of wine and is sitting there and grasses are already
sitting there and grasses are already
sitting there and grasses are already poured and we’re all drinking happily I
poured and we’re all drinking happily I
poured and we’re all drinking happily I mean this is not how it normally works
mean this is not how it normally works
mean this is not how it normally works but what’s normal
but what’s normal
but what’s normal how do we define normal how do we do we
how do we define normal how do we do we
how do we define normal how do we do we find usual now I’ll just show you a
find usual now I’ll just show you a
find usual now I’ll just show you a couple of things and I know that I might
couple of things and I know that I might
couple of things and I know that I might actually travel on dangerous grounds and
actually travel on dangerous grounds and
actually travel on dangerous grounds and surf and dangerous waters but let’s
surf and dangerous waters but let’s
surf and dangerous waters but let’s let’s see how he goes now this is also a
let’s see how he goes now this is also a
let’s see how he goes now this is also a pretty general statement and you might
pretty general statement and you might
pretty general statement and you might find lots of counter examples probably
find lots of counter examples probably
find lots of counter examples probably not in my case but it’s clear actually
not in my case but it’s clear actually
not in my case but it’s clear actually if you search Google with this you will
if you search Google with this you will
if you search Google with this you will come up with almost all counter examples
come up with almost all counter examples
come up with almost all counter examples another general very general statement
another general very general statement
another general very general statement is this and it’s true
is this and it’s true
is this and it’s true it’s true I might behave a little
it’s true I might behave a little
it’s true I might behave a little tonight but I am Italian and I use my
tonight but I am Italian and I use my
tonight but I am Italian and I use my hands a lot when I talk so there is this
hands a lot when I talk so there is this
hands a lot when I talk so there is this is another big generalization yeah which
is another big generalization yeah which
is another big generalization yeah which is true to a point and how does it come
is true to a point and how does it come
is true to a point and how does it come about it come about by what we normally
about it come about by what we normally
about it come about by what we normally do and for what I’m interested in and
do and for what I’m interested in and
do and for what I’m interested in and I’m gonna tell you in a second it also
I’m gonna tell you in a second it also
I’m gonna tell you in a second it also come about by what we normally say and
come about by what we normally say and
come about by what we normally say and what we normally write because what I’m
what we normally write because what I’m
what we normally write because what I’m interested in is actually language and
interested in is actually language and
interested in is actually language and words and more specifically I’m
words and more specifically I’m
words and more specifically I’m interested in how we can actually teach
interested in how we can actually teach
interested in how we can actually teach machines to process the language that we
machines to process the language that we
machines to process the language that we speak and that we write like can we
speak and that we write like can we
speak and that we write like can we teach a machine to understand like
teach a machine to understand like
teach a machine to understand like English and Italian Dutch and so on I
English and Italian Dutch and so on I
English and Italian Dutch and so on I mean you probably all use Google
mean you probably all use Google
mean you probably all use Google Translate so that’s what I do is also
Translate so that’s what I do is also
Translate so that’s what I do is also behind that kind that kind of stuff and
behind that kind that kind of stuff and
behind that kind that kind of stuff and why does this have to do with
why does this have to do with
why does this have to do with generalizations well because to teach
generalizations well because to teach
generalizations well because to teach machines something they need to learn
machines something they need to learn
machines something they need to learn from very general patterns and in this
from very general patterns and in this
from very general patterns and in this specific context what I want to tell you
specific context what I want to tell you
specific context what I want to tell you about is teaching machine to do a very
about is teaching machine to do a very
about is teaching machine to do a very tricky generalization which is being
tricky generalization which is being
tricky generalization which is being able to tell from a piece of text who
able to tell from a piece of text who
able to tell from a piece of text who the author is but not just like
the author is but not just like
the author is but not just like specifically who the author is
specifically who the author is
specifically who the author is I want to profile the author so what
I want to profile the author so what
I want to profile the author so what does he mean profiling so you give me a
does he mean profiling so you give me a
does he mean profiling so you give me a text I give it to the Machine and I want
text I give it to the Machine and I want
text I give it to the Machine and I want the machine to be able to tell whether
the machine to be able to tell whether
the machine to be able to tell whether it was written say by a male or a female
it was written say by a male or a female
it was written say by a male or a female or a younger person or an older person
or a younger person or an older person
or a younger person or an older person we can also try to profile like even
we can also try to profile like even
we can also try to profile like even more difficult things for example
more difficult things for example
more difficult things for example personality and so but I’ll stick to
personality and so but I’ll stick to
personality and so but I’ll stick to gender and age especially gender
gender and age especially gender
gender and age especially gender and what is the underlying assumption
and what is the underlying assumption
and what is the underlying assumption for this and this is why generalizations
for this and this is why generalizations
for this and this is why generalizations are interesting so if I can teach a
are interesting so if I can teach a
are interesting so if I can teach a machine to tell whether a text was
machine to tell whether a text was
machine to tell whether a text was written by a female or a male that means
written by a female or a male that means
written by a female or a male that means that males and females write texts
that males and females write texts
that males and females write texts differently now is this true how many of
differently now is this true how many of
differently now is this true how many of you think this is true ha ha now what I
you think this is true ha ha now what I
you think this is true ha ha now what I did to start with there is a tool at the
did to start with there is a tool at the
did to start with there is a tool at the unit that was developed at the
unit that was developed at the
unit that was developed at the University of Trento in Italy that takes
University of Trento in Italy that takes
University of Trento in Italy that takes a very very large collection of tweets
a very very large collection of tweets
a very very large collection of tweets and you can just check what are frequent
and you can just check what are frequent
and you can just check what are frequent words used by male or females
words used by male or females
words used by male or females now again careful because this is
now again careful because this is
now again careful because this is something that could look a little
something that could look a little
something that could look a little dangerous let’s check one now this is
dangerous let’s check one now this is
dangerous let’s check one now this is the tool Twitter life now this is now
the tool Twitter life now this is now
the tool Twitter life now this is now made up yeah this is a collection of
made up yeah this is a collection of
made up yeah this is a collection of tweets and it looks like men tend to
tweets and it looks like men tend to
tweets and it looks like men tend to talk more about beer than women and
talk more about beer than women and
talk more about beer than women and women tend to talk more about tea than
women tend to talk more about tea than
women tend to talk more about tea than men now ha stereotypical is this now
men now ha stereotypical is this now
men now ha stereotypical is this now waste because it gets worse or better
waste because it gets worse or better
waste because it gets worse or better these are other words that I put in you
these are other words that I put in you
these are other words that I put in you can try a lot so you can just actually
can try a lot so you can just actually
can try a lot so you can just actually see how you go from ballet to volleyball
see how you go from ballet to volleyball
see how you go from ballet to volleyball to football to tech and now the gender
to football to tech and now the gender
to football to tech and now the gender balance is getting really skewed here so
balance is getting really skewed here so
balance is getting really skewed here so there’s another study that was done on
there’s another study that was done on
there’s another study that was done on Facebook and some researchers collected
Facebook and some researchers collected
Facebook and some researchers collected female pages on Facebook and Mail pages
female pages on Facebook and Mail pages
female pages on Facebook and Mail pages on Facebook and they built word clouds
on Facebook and they built word clouds
on Facebook and they built word clouds you probably have seen word clouds you
you probably have seen word clouds you
you probably have seen word clouds you can visualize the most frequent words
can visualize the most frequent words
can visualize the most frequent words like bigger and the less frequent words
like bigger and the less frequent words
like bigger and the less frequent words smaller now I’m gonna show you a cloud
smaller now I’m gonna show you a cloud
smaller now I’m gonna show you a cloud now that was built by some researchers
now that was built by some researchers
now that was built by some researchers on Facebook pages and you can try to
on Facebook pages and you can try to
on Facebook pages and you can try to raise your hand if you think this is a
raise your hand if you think this is a
raise your hand if you think this is a female cloud okay I’m gonna show it now
female cloud okay I’m gonna show it now
female cloud okay I’m gonna show it now how many of you think a female cloud
how many of you think a female cloud
how many of you think a female cloud there you go now I’m going to show you
there you go now I’m going to show you
there you go now I’m going to show you the male cloud but quick yeah and you
the male cloud but quick yeah and you
the male cloud but quick yeah and you probably see why I know I know I know
probably see why I know I know I know
probably see why I know I know I know and I swear this isn’t made up this is
and I swear this isn’t made up this is
and I swear this isn’t made up this is real data so it was actually curious
real data so it was actually curious
real data so it was actually curious because this is so stereotypical that
because this is so stereotypical that
because this is so stereotypical that it’s kind of scary and so I was really
it’s kind of scary and so I was really
it’s kind of scary and so I was really curious and I found out something that
curious and I found out something that
curious and I found out something that that I thought was actually interesting
that I thought was actually interesting
that I thought was actually interesting which is this I went to check how much
which is this I went to check how much
which is this I went to check how much the word stereotype is used by female
the word stereotype is used by female
the word stereotype is used by female email totally gender balanced I thought
email totally gender balanced I thought
email totally gender balanced I thought it was very interesting now why do we
it was very interesting now why do we
it was very interesting now why do we kind of use this information so if if
kind of use this information so if if
kind of use this information so if if it’s true that men and women use these
it’s true that men and women use these
it’s true that men and women use these different words in such a clear way then
different words in such a clear way then
different words in such a clear way then if I tell the Machine come on just look
if I tell the Machine come on just look
if I tell the Machine come on just look at the words that are used and I give it
at the words that are used and I give it
at the words that are used and I give it a lot of data to learn from it ill
a lot of data to learn from it ill
a lot of data to learn from it ill eventually come up with its own
eventually come up with its own
eventually come up with its own generalizations now mind you we are
generalizations now mind you we are
generalizations now mind you we are giving machine data we are not telling
giving machine data we are not telling
giving machine data we are not telling machine like look for beer or look for
machine like look for beer or look for
machine like look for beer or look for tea or look for tech we just give it a
tea or look for tech we just give it a
tea or look for tech we just give it a lot of data and when the machine will
lot of data and when the machine will
lot of data and when the machine will know this date is associated with mail
know this date is associated with mail
know this date is associated with mail this state is associated with female and
this state is associated with female and
this state is associated with female and then it will make its own
then it will make its own
then it will make its own generalizations okay now we don’t only
generalizations okay now we don’t only
generalizations okay now we don’t only use words we also use some other hints
use words we also use some other hints
use words we also use some other hints now for example one thing that you might
now for example one thing that you might
now for example one thing that you might not know what is the difference between
not know what is the difference between
not know what is the difference between these two
these two
these two yeah their nose yeah how many of you use
yeah their nose yeah how many of you use
yeah their nose yeah how many of you use that one with the nose ha now you might
that one with the nose ha now you might
that one with the nose ha now you might not know that at least in the social
not know that at least in the social
not know that at least in the social media world you count as old what we can
media world you count as old what we can
media world you count as old what we can tell a machine is because we don’t need
tell a machine is because we don’t need
tell a machine is because we don’t need to say you know look at specific words
to say you know look at specific words
to say you know look at specific words we can also say you know look at
we can also say you know look at
we can also say you know look at sequences of characters yeah so they can
sequences of characters yeah so they can
sequences of characters yeah so they can also spot some hidden patterns that we
also spot some hidden patterns that we
also spot some hidden patterns that we might not be able to see so
might not be able to see so
might not be able to see so straightforwardly now there was so if we
straightforwardly now there was so if we
straightforwardly now there was so if we if we teach this to machines are we
if we teach this to machines are we
if we teach this to machines are we teaching machines to recognize basically
teaching machines to recognize basically
teaching machines to recognize basically stereotypes as much as we seem to use
stereotypes as much as we seem to use
stereotypes as much as we seem to use now there was a study that was published
now there was a study that was published
now there was a study that was published last year at a major conference in
last year at a major conference in
last year at a major conference in computational linguistics and the people
computational linguistics and the people
computational linguistics and the people who did this study aimed at checking
who did this study aimed at checking
who did this study aimed at checking exactly this so they took a system and
exactly this so they took a system and
exactly this so they took a system and they took people and they made them
they took people and they made them
they took people and they made them assign gender to the same data now what
assign gender to the same data now what
assign gender to the same data now what they found they found actually three
they found they found actually three
they found they found actually three very interesting things so number one I
very interesting things so number one I
very interesting things so number one I have a quote from the paper which is
have a quote from the paper which is
have a quote from the paper which is this so humans they found out the humans
this so humans they found out the humans
this so humans they found out the humans relied on correct stereotypes but relied
relied on correct stereotypes but relied
relied on correct stereotypes but relied on them more heavily than warranted from
on them more heavily than warranted from
on them more heavily than warranted from the data what does it mean that they
the data what does it mean that they
the data what does it mean that they over generalized on the beer and the tea
over generalized on the beer and the tea
over generalized on the beer and the tea and the volleyball and all the tech and
and the volleyball and all the tech and
and the volleyball and all the tech and so on yeah so what happens here is that
so on yeah so what happens here is that
so on yeah so what happens here is that we are loaded with stereotypes and we
we are loaded with stereotypes and we
we are loaded with stereotypes and we actually use them in our evaluation of
actually use them in our evaluation of
actually use them in our evaluation of data so if I want to say if we ask
data so if I want to say if we ask
data so if I want to say if we ask around apparently this was like used too
around apparently this was like used too
around apparently this was like used too much the stereotypical information now
much the stereotypical information now
much the stereotypical information now so they make assumptions of that kind
so they make assumptions of that kind
so they make assumptions of that kind now the other thing though that they
now the other thing though that they
now the other thing though that they found out the second thing which is very
found out the second thing which is very
found out the second thing which is very interesting is that nevertheless humans
interesting is that nevertheless humans
interesting is that nevertheless humans can tell gender pretty well in about 75
can tell gender pretty well in about 75
can tell gender pretty well in about 75 80 percent of the cases
80 percent of the cases
80 percent of the cases how about machines approximately the
how about machines approximately the
how about machines approximately the same no machines can tell gender in
same no machines can tell gender in
same no machines can tell gender in approximate approximately 75 to 80
approximate approximately 75 to 80
approximate approximately 75 to 80 percent of the cases depending on
percent of the cases depending on
percent of the cases depending on language depending on the text and so on
language depending on the text and so on
language depending on the text and so on but the third thing that they found and
but the third thing that they found and
but the third thing that they found and it’s the most interesting one is that if
it’s the most interesting one is that if
it’s the most interesting one is that if you take a combined approach of humans
you take a combined approach of humans
you take a combined approach of humans and machines you get the highest
and machines you get the highest
and machines you get the highest accuracy what does this mean it means
accuracy what does this mean it means
accuracy what does this mean it means that machines and humans do not see the
that machines and humans do not see the
that machines and humans do not see the same things yeah you can think about it
same things yeah you can think about it
same things yeah you can think about it a bit like that yeah there’s a correct
a bit like that yeah there’s a correct
a bit like that yeah there’s a correct bunch that both machines and humans will
bunch that both machines and humans will
bunch that both machines and humans will see but at the sides that are correct
see but at the sides that are correct
see but at the sides that are correct cases that the humans will be able to
cases that the humans will be able to
cases that the humans will be able to judge and others that the machines will
judge and others that the machines will
judge and others that the machines will be able to judge well isn’t this
be able to judge well isn’t this
be able to judge well isn’t this interesting
interesting
interesting yeah you might not completely expect
yeah you might not completely expect
yeah you might not completely expect this now why does this happen this is
this now why does this happen this is
this now why does this happen this is because we do teach machines to look at
because we do teach machines to look at
because we do teach machines to look at words but we don’t say which words
words but we don’t say which words
words but we don’t say which words yeah they generalize over something that
yeah they generalize over something that
yeah they generalize over something that we might not see like for example do you
we might not see like for example do you
we might not see like for example do you know did you know that women apparently
know did you know that women apparently
know did you know that women apparently tend to use more pronouns than men now
tend to use more pronouns than men now
tend to use more pronouns than men now if you don’t know this because you’re
if you don’t know this because you’re
if you don’t know this because you’re not involved in this kind of in this
not involved in this kind of in this
not involved in this kind of in this field why would you start to look at the
field why would you start to look at the
field why would you start to look at the text you know what is more pronouns
text you know what is more pronouns
text you know what is more pronouns maybe it’s wrong but the machine will
maybe it’s wrong but the machine will
maybe it’s wrong but the machine will learn this or sequences of characters
learn this or sequences of characters
learn this or sequences of characters your specific sequences of characters
your specific sequences of characters
your specific sequences of characters that might be typical of a younger or of
that might be typical of a younger or of
that might be typical of a younger or of an older person yeah we might not notice
an older person yeah we might not notice
an older person yeah we might not notice them when a machine we learn now I want
them when a machine we learn now I want
them when a machine we learn now I want to show you another example that doesn’t
to show you another example that doesn’t
to show you another example that doesn’t have to do with language but it’s a very
have to do with language but it’s a very
have to do with language but it’s a very very very related thing so now do you
very very related thing so now do you
very very related thing so now do you know what this is yeah you’ve seen it
know what this is yeah you’ve seen it
know what this is yeah you’ve seen it mucinous now this is a match between
mucinous now this is a match between
mucinous now this is a match between alphago which is a system the place goal
alphago which is a system the place goal
alphago which is a system the place goal made by Google and léa Seydoux champion
made by Google and léa Seydoux champion
made by Google and léa Seydoux champion super champion now the hardest match
super champion now the hardest match
super champion now the hardest match Sato lost not all of the games but four
Sato lost not all of the games but four
Sato lost not all of the games but four out of five now there was a point in a
out of five now there was a point in a
out of five now there was a point in a game where alphago did something really
game where alphago did something really
game where alphago did something really odd totally out of the usual completely
odd totally out of the usual completely
odd totally out of the usual completely unexpected move so it places little
unexpected move so it places little
unexpected move so it places little thingy somewhere where like you know no
thingy somewhere where like you know no
thingy somewhere where like you know no human would ever do this and the
human would ever do this and the
human would ever do this and the commentator actually a first confess
commentator actually a first confess
commentator actually a first confess later thought it’s a mistake went
later thought it’s a mistake went
later thought it’s a mistake went bonkers right it’s just completely wrong
bonkers right it’s just completely wrong
bonkers right it’s just completely wrong this is not a real artificial
this is not a real artificial
this is not a real artificial intelligence machine but later something
intelligence machine but later something
intelligence machine but later something happened right later he realized it did
happened right later he realized it did
happened right later he realized it did make sense yeah so at first what he says
make sense yeah so at first what he says
make sense yeah so at first what he says is this like is not a human move it’s
is this like is not a human move it’s
is this like is not a human move it’s not it’s totally totally out of place
not it’s totally totally out of place
not it’s totally totally out of place but later what he notices it makes sense
but later what he notices it makes sense
but later what he notices it makes sense and the reason why it makes sense is
and the reason why it makes sense is
and the reason why it makes sense is that because the patterns that this
that because the patterns that this
that because the patterns that this machine has learned are not necessarily
machine has learned are not necessarily
machine has learned are not necessarily openly in our view they might just not
openly in our view they might just not
openly in our view they might just not be there but they are there yeah and
be there but they are there yeah and
be there but they are there yeah and what I really like about this is what it
what I really like about this is what it
what I really like about this is what it goes on yeah because it says you know
goes on yeah because it says you know
goes on yeah because it says you know alphago learns also from itself so the
alphago learns also from itself so the
alphago learns also from itself so the machine improves but people also improve
machine improves but people also improve
machine improves but people also improve humans improve because they learn from
humans improve because they learn from
humans improve because they learn from what actually the choices that the
what actually the choices that the
what actually the choices that the machine does yeah and then this is what
machine does yeah and then this is what
machine does yeah and then this is what how it continues you know the
how it continues you know the
how it continues you know the commentator is reporting this other guy
commentator is reporting this other guy
commentator is reporting this other guy saying you know that makes him happy no
saying you know that makes him happy no
saying you know that makes him happy no so beautiful he says this is so
so beautiful he says this is so
so beautiful he says this is so beautiful and I totally agree with that
beautiful and I totally agree with that
beautiful and I totally agree with that come this is really beautiful yeah we
come this is really beautiful yeah we
come this is really beautiful yeah we might be actually helped by machines to
might be actually helped by machines to
might be actually helped by machines to even overcome some of our normal
even overcome some of our normal
even overcome some of our normal thinking usual thinking possibly even
thinking usual thinking possibly even
thinking usual thinking possibly even stereotypical thinking yeah we have a
stereotypical thinking yeah we have a
stereotypical thinking yeah we have a chance so I was thinking about this
chance so I was thinking about this
chance so I was thinking about this thing in terms of rise and fall and I
thing in terms of rise and fall and I
thing in terms of rise and fall and I know that lots of people are kind of
know that lots of people are kind of
know that lots of people are kind of scared
scared
scared by AI taking over like a mushy snow
by AI taking over like a mushy snow
by AI taking over like a mushy snow alphago beating the man
alphago beating the man
alphago beating the man this is really bad he’s not bad it’s a
this is really bad he’s not bad it’s a
this is really bad he’s not bad it’s a chance it’s a it’s a fantastic
chance it’s a it’s a fantastic
chance it’s a it’s a fantastic opportunity that we have and this is it
opportunity that we have and this is it
opportunity that we have and this is it this is a piece on the Economist there
this is a piece on the Economist there
this is a piece on the Economist there was actually talking about this like you
was actually talking about this like you
was actually talking about this like you know rise of the machines artificial
know rise of the machines artificial
know rise of the machines artificial intelligence scares people and they say
intelligence scares people and they say
intelligence scares people and they say excessively so we shouldn’t get scared
excessively so we shouldn’t get scared
excessively so we shouldn’t get scared we should welcome this yeah we make
we should welcome this yeah we make
we should welcome this yeah we make these machines I mean we program them
these machines I mean we program them
these machines I mean we program them and we can get a lot out of them not
and we can get a lot out of them not
and we can get a lot out of them not only in terms of like how useful they
only in terms of like how useful they
only in terms of like how useful they are but also how much they can teach us
are but also how much they can teach us
are but also how much they can teach us even in creativity you know mind you
even in creativity you know mind you
even in creativity you know mind you they’re also they’re very powerful but
they’re also they’re very powerful but
they’re also they’re very powerful but there are things that they are not very
there are things that they are not very
there are things that they are not very good at now for example there’s this is
good at now for example there’s this is
good at now for example there’s this is a tweet and if you read it I’m pretty
a tweet and if you read it I’m pretty
a tweet and if you read it I’m pretty sure you understand why I put it there
yeah now this is sarcastic tweet and
yeah now this is sarcastic tweet and it’s something that for a machine is
it’s something that for a machine is
it’s something that for a machine is still extremely hard extremely hard to
still extremely hard extremely hard to
still extremely hard extremely hard to get because it relies on some knowledge
get because it relies on some knowledge
get because it relies on some knowledge of the world that the machine has
of the world that the machine has
of the world that the machine has difficulties in acquiring it doesn’t
difficulties in acquiring it doesn’t
difficulties in acquiring it doesn’t mean that machines are not creative
mean that machines are not creative
mean that machines are not creative apart from the alphago match which i
apart from the alphago match which i
apart from the alphago match which i find extremely fascinating there’s also
find extremely fascinating there’s also
find extremely fascinating there’s also a really recent experiment that I saw
a really recent experiment that I saw
a really recent experiment that I saw the other day on the web and there’s a
the other day on the web and there’s a
the other day on the web and there’s a researcher who trained a machine to
researcher who trained a machine to
researcher who trained a machine to produce new names for color patterns for
produce new names for color patterns for
produce new names for color patterns for color patterns right and then after
color patterns right and then after
color patterns right and then after several iterations of this learning some
several iterations of this learning some
several iterations of this learning some really really funny names come up and
really really funny names come up and
really really funny names come up and these are not names that we would come
these are not names that we would come
these are not names that we would come up with ourselves because they’re art
up with ourselves because they’re art
up with ourselves because they’re art but come on I put some here so yes now
but come on I put some here so yes now
but come on I put some here so yes now there’s even seen this poop at the end
there’s even seen this poop at the end
there’s even seen this poop at the end and light of blast and I have no stomach
and light of blast and I have no stomach
and light of blast and I have no stomach beige and sink I mean these are really
beige and sink I mean these are really
beige and sink I mean these are really kind of odd names but if you think about
kind of odd names but if you think about
kind of odd names but if you think about it you know why not
it you know why not
it you know why not we couldn’t ever probably we wouldn’t
we couldn’t ever probably we wouldn’t
we couldn’t ever probably we wouldn’t come up with this ourselves but why
come up with this ourselves but why
come up with this ourselves but why should we get like afraid of embracing
should we get like afraid of embracing
should we get like afraid of embracing actually some additional creativity that
actually some additional creativity that
actually some additional creativity that machine’s can can give us yeah it’s it’s
machine’s can can give us yeah it’s it’s
machine’s can can give us yeah it’s it’s a different point of view we should
a different point of view we should
a different point of view we should welcome this we can learn a lot and why
welcome this we can learn a lot and why
welcome this we can learn a lot and why not we shouldn’t get we shouldn’t get
not we shouldn’t get we shouldn’t get
not we shouldn’t get we shouldn’t get scared at all so I really hope that we
scared at all so I really hope that we
scared at all so I really hope that we can welcome machines not just as useful
can welcome machines not just as useful
can welcome machines not just as useful tools but also to provide a point of
tools but also to provide a point of
tools but also to provide a point of view that is potentially a little
view that is potentially a little
view that is potentially a little different than what we are stuck with at
different than what we are stuck with at
different than what we are stuck with at times and of course this is also
times and of course this is also
times and of course this is also something that you can do yourself right
something that you can do yourself right
something that you can do yourself right it’s not just the machines you can just
it’s not just the machines you can just
it’s not just the machines you can just once in a while introduce something
once in a while introduce something
once in a while introduce something different in your lives just change the
different in your lives just change the
different in your lives just change the way you view things like you know yeah I
way you view things like you know yeah I
way you view things like you know yeah I really have to thank my students for
really have to thank my students for
really have to thank my students for what they brought the other day it was
what they brought the other day it was
what they brought the other day it was such a surprise and if this is not just
such a surprise and if this is not just
such a surprise and if this is not just a thank you for the wine it’s more a
a thank you for the wine it’s more a
a thank you for the wine it’s more a thing which was nice and it was a very
thing which was nice and it was a very
thing which was nice and it was a very productive meeting but it’s more I thank
productive meeting but it’s more I thank
productive meeting but it’s more I thank you for for the novelty for for breaking
you for for the novelty for for breaking
you for for the novelty for for breaking the scheme’s for not just always
the scheme’s for not just always
the scheme’s for not just always sticking to the same meetings you know
sticking to the same meetings you know
sticking to the same meetings you know that’s not something different we can do
that’s not something different we can do
that’s not something different we can do it ourselves
it ourselves
it ourselves but we can also you know get some help
but we can also you know get some help
but we can also you know get some help from machines to introduce some novelty
from machines to introduce some novelty
from machines to introduce some novelty and some unexpected in our lives thank
and some unexpected in our lives thank
and some unexpected in our lives thank you
you
you you
”
Be First to Comment