fun computing it’s it’s very big news
these days so if you know if you’re
familiar with you know companies like
Google IBM Intel they’re all investing
very very heavily into developing what’s
called a universal quantum computer and
their idea is that you will be able to
do things that you cannot do with a
classical computer but if you were to
just kind of look take a look around
scan the headlines and see what the
current state of the field is you could
be forgiven for being a little bit
confused so I’m just gonna take a
sampling of some headlines that I’ve
found recently so okay the era of
quantum computing is here outlook cloudy
okay so is it here is it not what does
this what does this headline actually
mean what are quantum computers and how
do they work I think that’s that’s great
I hope to educate you a little bit about
that today
making quantum heaters reliable again
that’s good but what does that mean
there are they unreliable what you know
what what does it mean when we say
quality over quantity and this in this
context Google moves towards quantum
supremacy with a 72 cubic computer well
that sounds like we’re really making
progress like really quickly I mean
quantum supremacy sounds sounds pretty
cool and again we can ask the important
questions in life you know what what
happens to my bitcoins when a quantum
computer comes around the black but we
might be getting a little bit ahead of
ourselves there because you also will
find headlines like this how close are
we really to building a quantum computer
and that makes it sound like it’s not
such a so rosy of an outlook and then
this one I found the promise of quantum
computing supremacy is bunk which means
that this is never going to happen so
you know what is it how are we like
building a quantum computer and yet
the same time we have other articles
saying that this is never going to
happen what is like and the fact is both
of these things are actually true
so let me explain how this could
possibly be true and to do that
unfortunately I need you to teach you a
little bit of quantum mechanics first so
let’s talk about what a qubit actually
is so there are many many ways to make a
qubit in fact nobody has even agreed
what system a qubit is going to be in
the future but let me take a very simple
example because I’m sure you have all
familiar with this system and it is an
electron so you know from your high
school chemistry class that lectrons
have a mass and a charge that’s kind of
what makes them an electron and they’re
all identical
as you delve deeper electrons also have
what’s called an angular momentum which
just means you can treat them like
they’re spinning and that’s actually
what we call this we just call it spin
and so when you have charge that’s
spinning around in a circle what you get
is a magnetic field so basically an
electron is like the little tiniest
magnet that you can have and so what
happens now is if I put that electron in
another magnetic field like some
external field what happens is that
electron they will either it the
electrons magnetic field will either
align parallel or what’s called anti
parallel so aligned against it and
there’s an energy difference between
those it takes more energy to be aligned
anti parallel than it does parallel and
that’s cool because we can label these
two states will call them zero and one
and those will correspond to like our
you know our binary computer states 0
and 1 for our computations and how we
switch between them as I just send in a
small pulse of light and I can flip my
electron from spin down to spin up or 0
to 1 whatever you want to call it so ok
great I’ve got a two-level system that I
can flip back and forth between I can
control it individually ok now where
does the quantumness come in ok now this
is the part where things start to get a
little bit strange so I said on the last
slide I’ve got an electron it’s in the
zero
State I can send in a pulse of light and
I can flip it to the one state now what
happens if I send in a pulse of light
that is only half as strong does it
ignore the pulse does it absorb the
pulse does it go to some weird
in-between state what happens well bear
with me here there is a 50% chance that
the electron absorbs the pulse of light
and flips to the one state and there’s a
50% chance that nothing happens and it
stays in the zero state and we don’t
know what state it’s in until we look at
it and so what we call that when it’s in
this in-between state we call that a
superposition the electron is basically
in the up-and-down state at the same
time this is just kind of a mathematical
way of how we write this so that’s
that’s all it just means it’s in a
combination of up and down now if I take
this state and I measure it I can only
ever get 0 or 1 it has to return one of
those two states so how do I know that
it’s actually in between how do I know
that it’s in a 50/50 state well I repeat
this many many times
and if I do this many times over I’ll
find that half the time it ends up in
the spin up state and half the time it
ends in the spin down state and so this
is sort of like flipping a coin I have a
coin I can’t tell you what the next flip
is going to be but I know it’s gonna be
one of two answers and every if I flip
the coin many many times I’m going to
get roughly half heads and roughly half
tails so that’s kind of sort of what’s
happening here and this becomes more
more useful when I add more qubits so
this is where the magic comes in because
if I have two qubits or two electrons
together they cease to become one system
or sorry they cease to become two
systems and they become one system so
now they’re in a superposition
but if four states they can both be down
one can be up and down in either
combination and they can both be up so
if I were to add a third qubit I get
eight possible States
if I were to add a fourth cubit 16
states so it doubles with every cubit
you add and so 72 cubits as Google has
been working on recently is actually a
lot of computational power because I
have all of these states available to do
computations and that’s where the power
of quantum computing comes in is in the
superposition States now the problem is
that having these super positions alway
also means you have significant
downsides and these things are like used
to get like the the advantages and
disadvantages of quantum computing are
completely inseparable from each other
and so let’s talk about some of the
disadvantages for just a minute so let’s
say I’ve got an electron and it’s in my
experimental set up and I am working my
friendly quantum scientist he is
interacting with his system he is
putting that qubit into a superposition
he’s measuring it he’s having a lot of
fun trust me we have a lot of fun in the
lab doing this and it is not at all
tedious the problem is is that if I if
my system is open for me to be able to
measure it it also means that it’s open
to my environment and when I say the
environment I mean like heat light noise
in the electronics and what cat can do
sometimes is that can interfere with my
system so let’s say I’ve got this
superposition state prepared and I want
to do some computations but then some
random event the environment sneaks in
there and steals all of my information
before I get a chance to do anything
with it now I’ve got to start all over
again so maintaining these superposition
States is a big challenge and so here’s
an analogy I like to use so when I was
little I played this game called
perfection and so the idea was you have
this you had this plastic box and you
push this blue portion down and then you
turn the time and while the timer is
ticking down you’re racing you’re
scrambling to put of the
yellow pieces into their respective
respective places and as soon as the
timer runs down the blue portion pops up
and all the pieces you’ve carefully
prepared just go flying everywhere it’s
really frustrating I would not I would
not recommend it that’s what we had to
do to entertain ourselves before the
Internet so this is sort of what it’s
like doing quantum computation you’re
basically racing against the clock if
you want to do very complex calculations
chances are you’re not going to have
enough time so we call this the
coherence time and right now the
coherence time of qubits are not great
here’s another problem errors so yeah a
couple slides ago this slide will look
very familiar I said that you can send
in a pulse of light and you can flip a
qubit from zero to one now what happens
if your either your qubit or your pulse
of light are not perfect they’re maybe
99% perfect that’s pretty good right
well what happens if it’s 99% perfect
means that there’s a 99% chance that the
qubit flips to the Upstate but there’s
also a 1% chance that the qubit does
nothing so in this case one out of a
hundred times you’ve actually done the
opposite of what you want to do and so
instead of just having a perfect flip
from zero to one you’ve now created
another superposition state so which
means that when I measure this 99 times
out of 100 I’ll get the right state I’ll
get what I intended but it also means
that one time I’ll get something that I
didn’t intend and so my computer will
then give me the wrong answer and let me
tell you 99 percent is not great when
you compare it to computers nowadays the
error rate of your desktop computer or
the transistors in your iPhone is much
much much much lower than that like
maybe one in 10 to the 20 or something
like that not 99 percent one percent but
the fact is this is inevitable this is
an inevitable consequence
of superposition of quantum mechanics
and so the struggle has become using the
advantages versus mitigating the
disadvantages and so this is a this is a
graphic put together by IBM and so let
me just explain the important parts to
you because there’s some probably small
text on there the axis along this way is
the number of qubits in the processor
and the axis along this way is the error
rate how reliable your computer is or
your qubit is excuse me and the size of
the blue cube represents your
computational power and so what this is
basically saying is if I go along this
axis and I just keep adding more qubits
I don’t actually get any increase in
computational power because I’m just
basically adding more errors so unless I
can simultaneously increase the number
of qubits well decreasing the error rate
I’m never going to get a powerful enough
quantum computer to overtake classical
computers and so this is what makes
progress so hard to quantify sometimes
our 50 cubits or 72 qubits better I
don’t know it depends on what the error
rate is will we ever get an error rate
lower enough low enough to to fully kind
of unlock the power of these some people
say no and that’s why people argue that
this can never this can never happen we
can never get a universal quantum
computer and so this is this is the
debate this is the progress and there
are lots of people that are working on
both of these areas but right now we’re
very very limited in what quantum
computers can do so I just leave you
with this if that sounds a little bit
grim let me give you a little bit of
hope so there was once a time when
classical computers were in a state like
this so this one on the left is what’s
called ENIAC was built by US government
and it costs was at 6.8 million dollars
and it was the size of a room and now
IBM can make chips smaller than a grain
of salt for about 10 cents apiece
and this was about a seventy year gap
between these so right now experimental
quantum computing is maybe fifteen to
twenty years old so maybe we need
another 50 years maybe we need another
70 years this is certainly not something
the point where quantum computers
overtake every day computers is
certainly not right around the corner
but I am hopeful nonetheless so are they
the future yes I would say so eventually
but maybe not your future maybe like
your children or your grandchildren’s
future thank you [Applause]
these days so if you know if you’re
familiar with you know companies like
Google IBM Intel they’re all investing
very very heavily into developing what’s
called a universal quantum computer and
their idea is that you will be able to
do things that you cannot do with a
classical computer but if you were to
just kind of look take a look around
scan the headlines and see what the
current state of the field is you could
be forgiven for being a little bit
confused so I’m just gonna take a
sampling of some headlines that I’ve
found recently so okay the era of
quantum computing is here outlook cloudy
okay so is it here is it not what does
this what does this headline actually
mean what are quantum computers and how
do they work I think that’s that’s great
I hope to educate you a little bit about
that today
making quantum heaters reliable again
that’s good but what does that mean
there are they unreliable what you know
what what does it mean when we say
quality over quantity and this in this
context Google moves towards quantum
supremacy with a 72 cubic computer well
that sounds like we’re really making
progress like really quickly I mean
quantum supremacy sounds sounds pretty
cool and again we can ask the important
questions in life you know what what
happens to my bitcoins when a quantum
computer comes around the black but we
might be getting a little bit ahead of
ourselves there because you also will
find headlines like this how close are
we really to building a quantum computer
and that makes it sound like it’s not
such a so rosy of an outlook and then
this one I found the promise of quantum
computing supremacy is bunk which means
that this is never going to happen so
you know what is it how are we like
building a quantum computer and yet
the same time we have other articles
saying that this is never going to
happen what is like and the fact is both
of these things are actually true
so let me explain how this could
possibly be true and to do that
unfortunately I need you to teach you a
little bit of quantum mechanics first so
let’s talk about what a qubit actually
is so there are many many ways to make a
qubit in fact nobody has even agreed
what system a qubit is going to be in
the future but let me take a very simple
example because I’m sure you have all
familiar with this system and it is an
electron so you know from your high
school chemistry class that lectrons
have a mass and a charge that’s kind of
what makes them an electron and they’re
all identical
as you delve deeper electrons also have
what’s called an angular momentum which
just means you can treat them like
they’re spinning and that’s actually
what we call this we just call it spin
and so when you have charge that’s
spinning around in a circle what you get
is a magnetic field so basically an
electron is like the little tiniest
magnet that you can have and so what
happens now is if I put that electron in
another magnetic field like some
external field what happens is that
electron they will either it the
electrons magnetic field will either
align parallel or what’s called anti
parallel so aligned against it and
there’s an energy difference between
those it takes more energy to be aligned
anti parallel than it does parallel and
that’s cool because we can label these
two states will call them zero and one
and those will correspond to like our
you know our binary computer states 0
and 1 for our computations and how we
switch between them as I just send in a
small pulse of light and I can flip my
electron from spin down to spin up or 0
to 1 whatever you want to call it so ok
great I’ve got a two-level system that I
can flip back and forth between I can
control it individually ok now where
does the quantumness come in ok now this
is the part where things start to get a
little bit strange so I said on the last
slide I’ve got an electron it’s in the
zero
State I can send in a pulse of light and
I can flip it to the one state now what
happens if I send in a pulse of light
that is only half as strong does it
ignore the pulse does it absorb the
pulse does it go to some weird
in-between state what happens well bear
with me here there is a 50% chance that
the electron absorbs the pulse of light
and flips to the one state and there’s a
50% chance that nothing happens and it
stays in the zero state and we don’t
know what state it’s in until we look at
it and so what we call that when it’s in
this in-between state we call that a
superposition the electron is basically
in the up-and-down state at the same
time this is just kind of a mathematical
way of how we write this so that’s
that’s all it just means it’s in a
combination of up and down now if I take
this state and I measure it I can only
ever get 0 or 1 it has to return one of
those two states so how do I know that
it’s actually in between how do I know
that it’s in a 50/50 state well I repeat
this many many times
and if I do this many times over I’ll
find that half the time it ends up in
the spin up state and half the time it
ends in the spin down state and so this
is sort of like flipping a coin I have a
coin I can’t tell you what the next flip
is going to be but I know it’s gonna be
one of two answers and every if I flip
the coin many many times I’m going to
get roughly half heads and roughly half
tails so that’s kind of sort of what’s
happening here and this becomes more
more useful when I add more qubits so
this is where the magic comes in because
if I have two qubits or two electrons
together they cease to become one system
or sorry they cease to become two
systems and they become one system so
now they’re in a superposition
but if four states they can both be down
one can be up and down in either
combination and they can both be up so
if I were to add a third qubit I get
eight possible States
if I were to add a fourth cubit 16
states so it doubles with every cubit
you add and so 72 cubits as Google has
been working on recently is actually a
lot of computational power because I
have all of these states available to do
computations and that’s where the power
of quantum computing comes in is in the
superposition States now the problem is
that having these super positions alway
also means you have significant
downsides and these things are like used
to get like the the advantages and
disadvantages of quantum computing are
completely inseparable from each other
and so let’s talk about some of the
disadvantages for just a minute so let’s
say I’ve got an electron and it’s in my
experimental set up and I am working my
friendly quantum scientist he is
interacting with his system he is
putting that qubit into a superposition
he’s measuring it he’s having a lot of
fun trust me we have a lot of fun in the
lab doing this and it is not at all
tedious the problem is is that if I if
my system is open for me to be able to
measure it it also means that it’s open
to my environment and when I say the
environment I mean like heat light noise
in the electronics and what cat can do
sometimes is that can interfere with my
system so let’s say I’ve got this
superposition state prepared and I want
to do some computations but then some
random event the environment sneaks in
there and steals all of my information
before I get a chance to do anything
with it now I’ve got to start all over
again so maintaining these superposition
States is a big challenge and so here’s
an analogy I like to use so when I was
little I played this game called
perfection and so the idea was you have
this you had this plastic box and you
push this blue portion down and then you
turn the time and while the timer is
ticking down you’re racing you’re
scrambling to put of the
yellow pieces into their respective
respective places and as soon as the
timer runs down the blue portion pops up
and all the pieces you’ve carefully
prepared just go flying everywhere it’s
really frustrating I would not I would
not recommend it that’s what we had to
do to entertain ourselves before the
Internet so this is sort of what it’s
like doing quantum computation you’re
basically racing against the clock if
you want to do very complex calculations
chances are you’re not going to have
enough time so we call this the
coherence time and right now the
coherence time of qubits are not great
here’s another problem errors so yeah a
couple slides ago this slide will look
very familiar I said that you can send
in a pulse of light and you can flip a
qubit from zero to one now what happens
if your either your qubit or your pulse
of light are not perfect they’re maybe
99% perfect that’s pretty good right
well what happens if it’s 99% perfect
means that there’s a 99% chance that the
qubit flips to the Upstate but there’s
also a 1% chance that the qubit does
nothing so in this case one out of a
hundred times you’ve actually done the
opposite of what you want to do and so
instead of just having a perfect flip
from zero to one you’ve now created
another superposition state so which
means that when I measure this 99 times
out of 100 I’ll get the right state I’ll
get what I intended but it also means
that one time I’ll get something that I
didn’t intend and so my computer will
then give me the wrong answer and let me
tell you 99 percent is not great when
you compare it to computers nowadays the
error rate of your desktop computer or
the transistors in your iPhone is much
much much much lower than that like
maybe one in 10 to the 20 or something
like that not 99 percent one percent but
the fact is this is inevitable this is
an inevitable consequence
of superposition of quantum mechanics
and so the struggle has become using the
advantages versus mitigating the
disadvantages and so this is a this is a
graphic put together by IBM and so let
me just explain the important parts to
you because there’s some probably small
text on there the axis along this way is
the number of qubits in the processor
and the axis along this way is the error
rate how reliable your computer is or
your qubit is excuse me and the size of
the blue cube represents your
computational power and so what this is
basically saying is if I go along this
axis and I just keep adding more qubits
I don’t actually get any increase in
computational power because I’m just
basically adding more errors so unless I
can simultaneously increase the number
of qubits well decreasing the error rate
I’m never going to get a powerful enough
quantum computer to overtake classical
computers and so this is what makes
progress so hard to quantify sometimes
our 50 cubits or 72 qubits better I
don’t know it depends on what the error
rate is will we ever get an error rate
lower enough low enough to to fully kind
of unlock the power of these some people
say no and that’s why people argue that
this can never this can never happen we
can never get a universal quantum
computer and so this is this is the
debate this is the progress and there
are lots of people that are working on
both of these areas but right now we’re
very very limited in what quantum
computers can do so I just leave you
with this if that sounds a little bit
grim let me give you a little bit of
hope so there was once a time when
classical computers were in a state like
this so this one on the left is what’s
called ENIAC was built by US government
and it costs was at 6.8 million dollars
and it was the size of a room and now
IBM can make chips smaller than a grain
of salt for about 10 cents apiece
and this was about a seventy year gap
between these so right now experimental
quantum computing is maybe fifteen to
twenty years old so maybe we need
another 50 years maybe we need another
70 years this is certainly not something
the point where quantum computers
overtake every day computers is
certainly not right around the corner
but I am hopeful nonetheless so are they
the future yes I would say so eventually
but maybe not your future maybe like
your children or your grandchildren’s
future thank you [Applause]
Please follow and like us: