WEBVTT

1
00:00:00.000 --> 00:00:05.099
Silence.

2
00:00:05.099 --> 00:00:12.718
Silence.

3
00:00:14.548 --> 00:00:25.469
Silence.

4
00:00:27.719 --> 00:00:32.759
Silence.

5
00:00:42.240 --> 00:00:58.590
Silence.

6
00:01:06.060 --> 00:01:10.799
Okay, good afternoon. Everyone this is.

7
00:01:10.799 --> 00:01:16.709
Class 14 quantum computer programming and.

8
00:01:16.709 --> 00:01:20.010
Again, just to make an audio check.

9
00:01:22.290 --> 00:01:35.665
Are you okay, so I have a chat. Hey, thank you. Just so I have a chat window open to my side. So, in theory you can post questions at any time.

10
00:01:35.665 --> 00:01:39.474
Don't wait for the end of anything. We can make things interactive. I'm not.

11
00:01:40.019 --> 00:01:52.349
Trying to wake people up by forcing you to ask questions, but you're certainly welcome to now what is happening today if the hardware works.

12
00:01:53.519 --> 00:01:58.590
And see here.

13
00:02:01.170 --> 00:02:04.200
Cool.

14
00:02:06.420 --> 00:02:10.110
Hey, wow. It actually sort of works, um.

15
00:02:10.405 --> 00:02:24.865
Really? It's working well, continuing on talking about hardware. So look at how this course has progressed doing is like, the progression of the history also doing the theory.

16
00:02:24.865 --> 00:02:37.585
1st and some algorithms because the theory and the allegations were developed there weren't real quantum computers. And now we're getting into the hardware of how they're actually being built and the field is still young.

17
00:02:37.615 --> 00:02:40.645
So that there are several competing different types of hardware.

18
00:02:41.909 --> 00:02:52.020
It will possibly shake out and some time, and the 1 type of hardware will become so much better than the others that you won't hear about the others. But we're still in this initial stage.

19
00:02:52.020 --> 00:03:00.719
At the start, so I had some general slides videos last time. So running through what.

20
00:03:00.719 --> 00:03:05.789
So, today I'm going to show you videos because or they're showing your stuff. I cannot easily talk about.

21
00:03:05.789 --> 00:03:11.969
And I got props and sides and so on and you never know suddenly much better speakers. And I am.

22
00:03:11.969 --> 00:03:15.449
And so we're going to have something talking about.

23
00:03:15.449 --> 00:03:19.949
Hardware something called a trend small and Cupid.

24
00:03:19.949 --> 00:03:31.740
And just a reminder in this video, they'll talk about things, the Hamel, Tony and and to oversimplify the Hamel Tony of a system is the.

25
00:03:31.740 --> 00:03:42.025
Is its energy so kinetic plus potential energy and it depends on that. Some for example, it's speed and kinetic energy. Depends on your speed.

26
00:03:42.025 --> 00:03:54.444
Potential energy depends on your position and actually for quantum mechanics. You've got Southerners equation and involves a hammer. Tony, and it talks about how the energy of the system depends on how these things.

27
00:03:54.750 --> 00:03:59.639
Change so if you're in a gravitational field, you get lower, you're.

28
00:03:59.639 --> 00:04:02.939
Potential energy decreases and so on so.

29
00:04:03.264 --> 00:04:17.995
That's what the Hamel telling you is, it's equation and you can derive the behavior of a system in terms of the Hamel, Tony and equation and derivatives. You can get different differential equations for how the energy depends on changes. And.

30
00:04:18.269 --> 00:04:23.579
Its position, its momentum and so on. And actually, this is.

31
00:04:23.579 --> 00:04:31.408
An equivalent way to describing the behavior of a system. Like, you've got noone's equations about.

32
00:04:31.408 --> 00:04:43.798
You know, velocity is conserved unless you have a force and so it's handled tony's alternative way to do that, which has certain advantages in certain cases and you can.

33
00:04:43.798 --> 00:04:52.858
Free to look it up. If you want certain advantages, it can handle constraints in a system the more easily perhaps in Newtonian.

34
00:04:52.858 --> 00:04:58.019
Um, if you got a ball rolling on a surface and you want to say, how does it's.

35
00:04:58.019 --> 00:05:11.428
How fast is it accelerate? It rolls down the curved inclined plane. Perhaps the Hamel, Tony, and in some cases might make it easier to do the math and a Newton than using Newton's equations directly because.

36
00:05:11.428 --> 00:05:21.059
You have to work in the constraints of the ball is on the incline plane, and perhaps the ball is not slipping and so on. Okay. So, and then we'll talk about.

37
00:05:21.564 --> 00:05:34.074
We've a long video here we'll see Alexander play will see part of it and now getting into more current stuff. And so these are very deep technical talks. I'll just point you to them.

38
00:05:34.403 --> 00:05:37.553
If you're interested the American Physical society.

39
00:05:38.249 --> 00:05:53.069
7 months ago, current, this year's meeting and I have the abstract included. Here's the whole session on it. So this 1, if you want to get into more detail, if you think I'm being too superficial with you, then you're free to look at that.

40
00:05:53.069 --> 00:06:07.379
And the next 1, I'll talk about, I'll show 10 minutes of the thing. So, in a sense, the cube, it's like artificial atoms and they've quantum property. So we'll see 10 minutes of that.

41
00:06:07.379 --> 00:06:19.408
Now, I showed you some videos on the Q tech academy in depth. So I hear I just have a link to their home page. So you can they got a lot of videos tutorial material.

42
00:06:19.408 --> 00:06:24.178
Okay, so then the next thing is, as there's time I'll talk about D, waves.

43
00:06:24.533 --> 00:06:34.764
Computer, which is a different process called quantum annealing so we've got a couple of videos on how D wave works and the last 1. I'll just show part of it.

44
00:06:34.764 --> 00:06:41.483
And then if you're just said, you can follow on on that, and then in addition to the videos, there are some.

45
00:06:43.019 --> 00:06:50.608
Got some slides here this, this page here actually has a started slides and.

46
00:06:50.608 --> 00:06:54.959
This is presented in principals of parallel computing conference a year ago.

47
00:06:56.783 --> 00:07:11.334
Okay, now so what's going to happen in future lecture next few lectures we're getting back. So now we've had hardware now we'll be getting back to some algorithm some applications and some got some nice description to that. So we'll go to the kiss.

48
00:07:11.334 --> 00:07:16.103
Could announce that page to talk about some algorithms and so on.

49
00:07:16.379 --> 00:07:21.899
And if you're still confused about quantum computing, Microsoft is a nice long.

50
00:07:21.899 --> 00:07:22.228
Um,

51
00:07:22.223 --> 00:07:25.163
video hour and a half long on now,

52
00:07:25.163 --> 00:07:25.853
the title's,

53
00:07:25.853 --> 00:07:27.653
the same as the textbook,

54
00:07:27.894 --> 00:07:29.514
but the speaker is different,

55
00:07:29.903 --> 00:07:33.923
but I think it presents things nicely and you,

56
00:07:33.983 --> 00:07:37.584
you're free if you still would like a more solid grounding.

57
00:07:37.584 --> 00:07:38.274
If you're a little.

58
00:07:39.028 --> 00:07:43.949
Uncertain at the foundations of your understanding, you might go to this.

59
00:07:43.949 --> 00:07:58.613
Book here any case, we'll see that another thing I've got some, another homework up and this is about several weeks ago. I walked you through programming the IBM queue. Now this homework is for you to do it.

60
00:07:58.613 --> 00:08:02.363
Yourself and this is from a tutorial that was at.

61
00:08:03.059 --> 00:08:09.778
Parallel programming principals conference and the idea is for you to read it and.

62
00:08:09.778 --> 00:08:13.228
Load and try some exercises and print out.

63
00:08:13.228 --> 00:08:22.048
The dogs just exercise was enough information to make it obvious that you could get it to work and just stop at this page here.

64
00:08:22.584 --> 00:08:32.783
So, you can go through, it's got slides. It's got examples. These are Python examples. So, watch the slides, create a kiss, get account.

65
00:08:32.813 --> 00:08:45.864
If you haven't and go through and then take this and run this thing, you might notice further down here. There's stuff about D wave and this might some of this might turn into next week's homework.

66
00:08:46.109 --> 00:08:56.578
So you'd be able to say on your resume that you programmed and if you program to D wave, and then at the end of this course, maybe you program Googles. So you'd be able to say.

67
00:08:56.578 --> 00:09:00.989
We programmed the big 3 or something. Okay.

68
00:09:00.989 --> 00:09:09.719
So, there's no questions so let's see a little about hardware here.

69
00:09:09.719 --> 00:09:18.028
And again, I'm taking it on trust that the audio will come through on this. If it doesn't end.

70
00:09:18.683 --> 00:09:32.994
Let me know, and I'm almost at the point, or I'm going to download the YouTube video separately without the ads, but I haven't quite had time to do that. Yet. Anyone wants to recommend a good ad blocker for you to.

71
00:09:33.298 --> 00:09:39.149
Yes, I would be glad to know about it.

72
00:09:39.149 --> 00:09:44.249
Silence.

73
00:09:44.249 --> 00:09:53.639
Silence.

74
00:09:53.639 --> 00:10:01.109
Getting the superconducting quantum circuits, or specifically quantum computing with transplant quantum bits or Cubans.

75
00:10:01.109 --> 00:10:11.428
In the circuit quantum electro dynamics, architecture superconducting Cubics differ in 2 important ways from truly quantum 2 level systems provided to us by nature.

76
00:10:11.428 --> 00:10:15.119
Such as the spin of electron or the span of certain.

77
00:10:15.119 --> 00:10:19.349
Like, carbon, 13 or Silicon 21st.

78
00:10:19.349 --> 00:10:23.458
There are multi level systems, not only the electronic levels in atoms.

79
00:10:23.458 --> 00:10:27.599
Multi level quantum systems can be used effectively as Cuba. It.

80
00:10:27.599 --> 00:10:30.869
By confining all dynamics to quantum levels.

81
00:10:30.869 --> 00:10:34.198
Usually the ground and the 1st, excited state of the system.

82
00:10:34.198 --> 00:10:38.788
2nd, these are circuits that we fabricate ourselves.

83
00:10:38.788 --> 00:10:44.129
This has advantages and disadvantages on the bright side. We have freedom to design.

84
00:10:44.129 --> 00:10:48.028
In a sense, we can play God designing artificial atoms.

85
00:10:48.028 --> 00:10:52.979
However, fabrication, uncertainty keeps us from making any the same.

86
00:10:52.979 --> 00:10:57.568
For this reason, you'd like to say that Cubans have individual personality.

87
00:10:57.568 --> 00:11:02.668
Well, the form of the describing them content mechanically is well known.

88
00:11:02.668 --> 00:11:07.288
The parameters of this Hamilton can not be perfectly targeted in fabrication.

89
00:11:07.288 --> 00:11:11.038
This is interesting implications for scaling up our quantum processors.

90
00:11:11.038 --> 00:11:14.428
Which I will discuss in a later video, but back to the design.

91
00:11:15.688 --> 00:11:22.438
Superconducting cubic generally consist of superconducting electrodes or islands that are interconnected by Joseph's injections.

92
00:11:22.438 --> 00:11:26.849
Tanya typically consists of 2 now commuting contributions.

93
00:11:26.849 --> 00:11:32.729
1, capacitive and 1 in the 1st turns to localize the Cooper pairs.

94
00:11:32.729 --> 00:11:36.448
And the 2nd favorite 3rd tunnelling from 1 island to another.

95
00:11:36.448 --> 00:11:42.328
In other words, they're helping their access numerous varieties of superconducting Cubans.

96
00:11:42.328 --> 00:11:47.129
The charged cubic flux, Cupid face cube it sodium.

97
00:11:47.129 --> 00:11:52.558
And many more these Cubans differ in terms of the number of superconducting islands.

98
00:11:52.558 --> 00:11:58.558
The number of junctions, and also quite importantly, the relative energy scales of the capacity and inducting.

99
00:11:59.818 --> 00:12:05.278
Typically superconducting Cuba it's working the frequency range between 4 and 8 gigahertz approximately.

100
00:12:05.278 --> 00:12:08.609
This frequency, let's call it 0 1.

101
00:12:08.609 --> 00:12:12.869
Is related via planck's constant to the energy difference. Easy 1.

102
00:12:12.869 --> 00:12:16.619
Between the quantum levels that we assign a state 0T and 1.

103
00:12:17.849 --> 00:12:24.448
0, 1 is also the frequency of the micro process that we will need to induce comparing transitions between levels.

104
00:12:25.979 --> 00:12:30.629
In this course, we will focus on the transport, a derivative of the charged cubic.

105
00:12:30.629 --> 00:12:35.009
The transplant is the superconducting Cuba that we specialize on the.

106
00:12:38.068 --> 00:12:42.989
In its simplest form, the transmen consists of 2 islands interconnected by 1 junction.

107
00:12:42.989 --> 00:12:47.639
Those of you familiar with circuits, particularly the electrical engineers.

108
00:12:47.639 --> 00:12:53.428
Will recognize that the transplant looks well, just like a parallel combination of 1 capacitor and 1 doctor.

109
00:12:53.428 --> 00:12:59.458
In other words, this gets us most of the way there. So let's take a look.

110
00:13:02.428 --> 00:13:07.739
The Hamilton for the so later consists of 2 terms that are each quite driving with respect to 1.

111
00:13:07.739 --> 00:13:12.178
Capacitive term is quadratic on the charge accumulated on 1 island.

112
00:13:12.178 --> 00:13:18.599
The opposite charges accumulator India the inductive term is quadratic on the flex to the doctor.

113
00:13:18.599 --> 00:13:24.298
This charge, and this flux do not commute. In fact, they are economically conjugate variables.

114
00:13:24.298 --> 00:13:30.568
A direct correspondence between this Hamilton can be made to that of the simple harmonic oscillator.

115
00:13:30.568 --> 00:13:33.899
By mapping flux to the position of the mass.

116
00:13:33.899 --> 00:13:46.408
And charge to the masses momentum. Physics students will not be surprised to learn that the spectrum of the month. I still see, also later is perfectly harmonic levels are equally spaced in energy.

117
00:13:46.408 --> 00:13:49.528
This equal spacing is given by a familiar formula.

118
00:13:49.528 --> 00:13:55.528
1, overrule LLC, but unfortunately a harmonic spectrum does not a good to make.

119
00:13:55.528 --> 00:13:59.399
It is very difficult to confine the dynamics to just 2 levels.

120
00:13:59.399 --> 00:14:03.538
So, leakage out of the cubic space is a permanent threat.

121
00:14:03.538 --> 00:14:08.458
That is why the transplant differs from the later in a fundamental way.

122
00:14:08.458 --> 00:14:12.869
In the transport inductance is provided by a Joseph's injunction.

123
00:14:12.869 --> 00:14:19.229
And not by a typical coil and doctor inductive energy for the junction is not quadratic.

124
00:14:19.229 --> 00:14:22.948
But, of course, I'm function of the generalized flux through it.

125
00:14:24.298 --> 00:14:32.188
This difference has important consequences for the spectrum. It disrupts the harmonic spectrum in ways that are very practical.

126
00:14:32.188 --> 00:14:36.928
At typical parameter values deep in the so called.

127
00:14:36.928 --> 00:14:41.458
Where the energy scale of the inductive term is much larger than that of the charging term.

128
00:14:41.458 --> 00:14:45.538
And let's say for a frequency of 0, 1 of about 6 gigahertz.

129
00:14:45.538 --> 00:14:51.869
The transition if 1, 2 from the 1st, to the 2nd excited state is lower by approximately.

130
00:14:51.869 --> 00:14:57.178
300 megahertz, this difference is sufficient in practice to confine the dynamics.

131
00:14:57.178 --> 00:15:02.099
So the 2 lowest levels, our cubic sub phase, when performing single cubic Gates.

132
00:15:02.099 --> 00:15:05.129
With policies of duration about 20 seconds.

133
00:15:05.129 --> 00:15:08.729
Brian will discuss the implementation of single cubic gates in detail.

134
00:15:10.229 --> 00:15:13.589
As a final note, we build our transplant not from 1.

135
00:15:13.589 --> 00:15:21.839
But 2, Joseph's injections in parallel, this gives us the possibility to tune the inductive element. And thereby also the Cuba transition frequency.

136
00:15:21.839 --> 00:15:26.068
By spreading a magnetic flux through the loop defined by the 2 junctions.

137
00:15:26.068 --> 00:15:30.778
We can do this independently for each Cuba and I'm now 2nd time tickets.

138
00:15:30.778 --> 00:15:34.318
This capability is the workforce enabling, took your big Gates.

139
00:15:34.318 --> 00:15:37.948
Other than I will present these in a later video.

140
00:15:37.948 --> 00:15:41.639
Sure.

141
00:15:50.548 --> 00:15:54.749
Watch this part of this, the start of this 1 here.

142
00:15:54.749 --> 00:15:57.989
Obviously, not all of it, but.

143
00:15:57.989 --> 00:16:08.009
Blog.

144
00:16:10.349 --> 00:16:18.028
Silence.

145
00:16:24.658 --> 00:16:29.249
So everyone and thanks for invitation to speak at this call.

146
00:16:29.249 --> 00:16:36.869
Um, so, 1st of all my goal is not to go through all of my slides, but my goal is for you guys to understand something. So.

147
00:16:36.869 --> 00:16:40.078
Please stop me anytime in the.

148
00:16:40.078 --> 00:16:45.448
Uh, the, the person's in front of me that I can go over time as much as I want anyway.

149
00:16:45.448 --> 00:16:49.798
So, it's fine, so please stop me at any time with with questions.

150
00:16:49.798 --> 00:16:54.869
So, I'm going to talk about using supervisory thing circuits for quantum information.

151
00:16:54.869 --> 00:17:00.089
Processing okay, so 1st of all, let's start very basic by something. You already know.

152
00:17:00.089 --> 00:17:05.009
Uh, but it's this will put the context around what I am going to talk about.

153
00:17:05.009 --> 00:17:09.209
So, what is the challenge here when we are thinking about.

154
00:17:09.209 --> 00:17:13.499
Permission processing and eventually building a quantum information processor.

155
00:17:13.499 --> 00:17:18.358
So 1st of all, we of course, and you all know that we need Cubans.

156
00:17:18.358 --> 00:17:22.709
You need quantum systems, which well defined.

157
00:17:22.709 --> 00:17:26.278
Levels 2 levels to family.

158
00:17:26.278 --> 00:17:31.858
And we would like to be able to control these Cubans with single unitary Gates.

159
00:17:31.858 --> 00:17:38.128
See, my expectations, for example, we of course need lots of these documents.

160
00:17:38.128 --> 00:17:44.429
I have a few minutes that don't talk to each other. That's boring. So we need these give us to interact with each other in some way or the other.

161
00:17:44.429 --> 00:17:48.058
To eventually and Tango these 2 bits of.

162
00:17:48.058 --> 00:17:52.469
Eventually all of these students, and at the end of the day, we need to.

163
00:17:52.469 --> 00:17:58.348
Re, out something from this computation thing, a bunch of zeros and ones which corresponds to the.

164
00:17:58.348 --> 00:18:04.949
And so to the population, this is, of course, is quite challenging because here we are asking for.

165
00:18:04.949 --> 00:18:08.009
Conflicting requirement 1st, we like these gates.

166
00:18:08.009 --> 00:18:13.769
To be super efficient, super fast. So of course quantum computers.

167
00:18:13.769 --> 00:18:18.479
Uh, are having benefits from exponential speed ups for certain task.

168
00:18:18.479 --> 00:18:22.048
But we still would like every gate to be relatively fast.

169
00:18:22.048 --> 00:18:26.999
And that means that need to be able to act on a cube very strongly.

170
00:18:26.999 --> 00:18:30.898
In the same way, we like this without process to be super efficient.

171
00:18:30.898 --> 00:18:35.038
We'd like to obtain information these zeroes and ones.

172
00:18:35.038 --> 00:18:39.808
Before they change before, for example, 1 turned into a 0T spontaneously.

173
00:18:39.808 --> 00:18:43.048
By the processes that you've heard about from the location yesterday.

174
00:18:43.048 --> 00:18:46.679
So, we would like our read out just to be strongly.

175
00:18:46.679 --> 00:18:50.489
To our Cubans, but at the same time, we'll want our cubes to be.

176
00:18:50.489 --> 00:18:54.659
Very coherent 1. is that all to be very sharp? We want this.

177
00:18:54.659 --> 00:18:58.348
This 0T in the position of 0T and wants to stay.

178
00:18:58.348 --> 00:19:02.878
Query and for a long time and so we would like our Cuba to be uncovered from everything.

179
00:19:02.878 --> 00:19:06.598
So this is the contracting these are the conflicting requirements.

180
00:19:06.598 --> 00:19:09.749
This is why this is difficult, but this is also why this is interesting.

181
00:19:09.749 --> 00:19:12.898
Okay, and in this doc.

182
00:19:12.898 --> 00:19:17.009
I will present to you 1 approach to realize this dream.

183
00:19:17.009 --> 00:19:22.108
Which is based on events, which are, will always also called artificial atoms.

184
00:19:22.108 --> 00:19:28.798
So, keywords, I can show that those have all systems to me. The same thing for me during this time.

185
00:19:28.798 --> 00:19:34.528
Okay, so I will start by saying you need a few words about potential ad downs. How can you.

186
00:19:34.528 --> 00:19:39.479
Build these artificial atoms, and if you build them other than the BA, actually in the app.

187
00:19:39.479 --> 00:19:45.509
Then I'll introduce an architecture circuit TV where you can hook up a bunch of these artificial atoms.

188
00:19:45.509 --> 00:19:48.749
And eventually do computation. I'll show.

189
00:19:48.749 --> 00:19:52.019
1, 1, such example of a.

190
00:19:52.019 --> 00:19:55.919
Simple competition, which has been realized with these artificial.

191
00:19:55.919 --> 00:20:03.479
And then I'll talk about future directions and challenges, although at that point, you probably the future direction you'll be.

192
00:20:03.479 --> 00:20:07.949
Thinking about this, probably lunch, but I'll pass your attention and.

193
00:20:07.949 --> 00:20:12.269
I'll show you where this is going. Okay so, artificial adjunct.

194
00:20:13.378 --> 00:20:24.058
So, 1st, let's review what we've already yesterday and, uh, Chris Monroe stock. So he talked about actual atoms. I'll call them here just to be super precise atomic atoms.

195
00:20:24.058 --> 00:20:29.669
That's fine for everyone. So we have artificial downs and we also have atomic atoms.

196
00:20:29.669 --> 00:20:35.878
And what is great is these atoms as loans have sharpening J levels.

197
00:20:35.878 --> 00:20:39.749
And it will say that 2 of these levels to be 0T than once.

198
00:20:39.749 --> 00:20:44.699
Okay, and something else, which is fantastic from these atoms.

199
00:20:44.699 --> 00:20:49.048
Is that their energy separations between the various levels? Is this thing.

200
00:20:49.048 --> 00:20:52.378
So, for example, the 0, 1 energy, a defense.

201
00:20:52.378 --> 00:20:56.159
Which I'm calling from 0T on here and frequency units is different.

202
00:20:56.159 --> 00:21:02.278
From Omega 1, 2 and what does that mean? But that means that if you want to control.

203
00:21:02.278 --> 00:21:05.459
The internal state of this.

204
00:21:05.459 --> 00:21:10.528
Using a laser for example, so just shine the bright laser as we've heard yesterday of the setup.

205
00:21:10.528 --> 00:21:13.919
And the typo moment as we again, yesterday.

206
00:21:13.919 --> 00:21:18.598
The type of moment of this Atom, a couple of to the electric sealed.

207
00:21:18.598 --> 00:21:26.068
Of the laser, and you will choose the laser to be at the 0, 1 condition frequency to the color.

208
00:21:26.068 --> 00:21:29.249
Of the laser, the frequency of the wave of these things are.

209
00:21:29.249 --> 00:21:32.338
To match this energy and in which case.

210
00:21:32.338 --> 00:21:37.888
You and use this, or if you have a sequence to sequence of navigate.

211
00:21:37.888 --> 00:21:43.378
You prefer this thing and because oh, my God 0, 1 is not equal to only got 1 2.

212
00:21:43.378 --> 00:21:49.648
There is very little leakage or no, the kids at all you stay in the 0T on some space and there's never the right.

213
00:21:49.648 --> 00:21:54.659
I mean, this laser is never providing the right energy for you to go from for the system to go from 1 to 2.

214
00:21:54.659 --> 00:21:58.919
So, you stay within the 0, 1 subspace. That's why this is a good.

215
00:21:58.919 --> 00:22:03.808
How good is it? Well, we agree that that 1 times.

216
00:22:03.808 --> 00:22:07.919
The T1 time you heard from Bill yesterday basically, if you can.

217
00:22:07.919 --> 00:22:12.028
The once they, how much time will it take to go back to the 0T States? Continuously?

218
00:22:12.028 --> 00:22:18.808
This can be a few years. I actually would like to see a measurement of this if you want to find at some point that.

219
00:22:18.808 --> 00:22:22.949
There's the smiling in the back. Yeah.

220
00:22:22.949 --> 00:22:28.108
We could use 2 or 3 post access to a lifetime to imagine this 1.

221
00:22:28.108 --> 00:22:31.199
Then, um.

222
00:22:31.199 --> 00:22:38.368
Which is much time with our superposition of 0T plus 1 state phase. What I'm going to say.

223
00:22:38.368 --> 00:22:42.239
Would say, 0T plus 1 before it goes to 0T minus 1, for example.

224
00:22:42.239 --> 00:22:46.199
That can be there are several seconds and that's fantastic.

225
00:22:46.199 --> 00:22:49.199
It's especially fantastic if you combine this with the fact that.

226
00:22:49.199 --> 00:22:54.328
Doing these gates preparing any civil position of 0T and 1 it's short except few microseconds.

227
00:22:54.328 --> 00:23:01.828
So, if you compare combine these 2 things, you'll find experimentally actually, from so called benchmarking protocol.

228
00:23:01.828 --> 00:23:05.249
You find that error brigade is small percent.

229
00:23:05.249 --> 00:23:08.759
So, you can do any single inhibit gates with very little error.

230
00:23:08.759 --> 00:23:12.358
And that's that's fantastic. That's why these make sense. I'm very fantastic for it.

231
00:23:12.358 --> 00:23:16.798
Quantum information processing. Okay. What about artificial atoms?

232
00:23:16.798 --> 00:23:20.189
Would like to use a standard toolkit of circuit.

233
00:23:20.189 --> 00:23:24.388
Standard circuit elements, and now create something.

234
00:23:24.388 --> 00:23:27.898
Which be AIDS, like an ambition like, it's like an.

235
00:23:27.898 --> 00:23:31.979
What is our tool kit? Well, we can use wires.

236
00:23:31.979 --> 00:23:35.219
Resisters and Victor and doctors and capacitors.

237
00:23:35.219 --> 00:23:39.598
Well, it shows all of these elements to be simple connecting.

238
00:23:39.598 --> 00:23:43.499
Well, let's forget about the register for the moment. Bluetooth. Our wires are inductors and capacitors.

239
00:23:43.499 --> 00:23:46.558
To be super connecting in a Super collector.

240
00:23:46.558 --> 00:23:50.788
Uh, you might know if you, if you pass a current to a regular wire.

241
00:23:50.788 --> 00:23:56.489
Then there is, it is dissipated. For example, this is all the bicarb spark.

242
00:23:56.489 --> 00:24:00.479
Passer current to a tiny wire and it limits.

243
00:24:00.479 --> 00:24:04.199
Energy in a simple conductor, there's no energy in it. It.

244
00:24:04.199 --> 00:24:08.788
There's no dissipation and intuitively that should be good for stuff.

245
00:24:08.788 --> 00:24:12.058
We use simple that.

246
00:24:12.058 --> 00:24:16.078
So, let's make the.

247
00:24:16.078 --> 00:24:20.818
Uh, Circuit out of this 10 of kit, and we'll take the parallel combination.

248
00:24:20.818 --> 00:24:26.338
Of an conductor and a capacity. So if there, if you don't remember this from.

249
00:24:26.338 --> 00:24:29.519
Undergrad or advanced high school.

250
00:24:29.519 --> 00:24:35.128
And that takes well, I'll tell you that the energy.

251
00:24:35.128 --> 00:24:38.429
Of such a 2nd, so the.

252
00:24:38.429 --> 00:24:41.818
Is the sum of 2 times the 1st time is.

253
00:24:41.818 --> 00:24:45.598
Charges charges are charging this capacity.

254
00:24:45.598 --> 00:24:51.148
This is a cool energy. You might remember that cool armies an electron electron traction.

255
00:24:51.148 --> 00:24:54.568
Put it in the square Square.

256
00:24:54.568 --> 00:24:59.548
Over to see, that's the size of this capacity and then there is a flux.

257
00:24:59.548 --> 00:25:03.179
Which is treading this.

258
00:25:03.179 --> 00:25:06.628
This is this this and Tucker sorry?

259
00:25:06.628 --> 00:25:11.278
If don't know what a flux is. Well, just take their voltage at the end points.

260
00:25:11.278 --> 00:25:16.769
Of this doctor and integrate overtime, this voltage that gives you the flux.

261
00:25:16.769 --> 00:25:21.058
So, you know what our voltage is, you know, what an integral is and you know what a flux is that's.

262
00:25:22.288 --> 00:25:26.338
That's the energy it goes as 5 square over to him.

263
00:25:26.338 --> 00:25:30.358
And I'm planning the energy the this time actually.

264
00:25:30.358 --> 00:25:33.568
As a function of flux and what you see, of course, is a parabola.

265
00:25:33.568 --> 00:25:36.598
And what we know from quantum mechanics is that.

266
00:25:36.598 --> 00:25:40.199
So this corresponds to a sort of harmony.

267
00:25:40.199 --> 00:25:46.019
Would you write about from those? Those aren't physicists you've heard about computers from the equation.

268
00:25:46.019 --> 00:25:49.499
We know also that the frequency of this, or when you consider.

269
00:25:49.499 --> 00:25:53.519
Will be given by 1 of our square root of else. That's the energy.

270
00:25:54.898 --> 00:25:59.038
Is that a good Cuban so far? There's very good news.

271
00:25:59.038 --> 00:26:05.759
You can choose by fabrication by using modern micro fabrication techniques.

272
00:26:05.759 --> 00:26:12.388
You can use, you can choose just at the frequencies of these us theaters is in the gig areas range.

273
00:26:12.388 --> 00:26:15.479
Various range, that's basically the frequencies of your.

274
00:26:15.479 --> 00:26:20.548
Cell phones and cell phones, there's an active element, which is basically announced circuit, which.

275
00:26:20.548 --> 00:26:24.778
Reacts at these migrate frequency.

276
00:26:24.778 --> 00:26:28.499
So, why is that a good news? Well, you can convert.

277
00:26:28.499 --> 00:26:33.659
Energy temperature and that from this, you see that and.

278
00:26:33.659 --> 00:26:40.019
Vigorous corresponds to Africa, but this is a pretty small temperature, but it's possible to operate.

279
00:26:40.019 --> 00:26:43.288
These circuits add something like 10 Calvin.

280
00:26:43.288 --> 00:26:46.739
Very easily using the fetch refrigerators.

281
00:26:46.739 --> 00:26:49.798
And so you will override the circuits will cool down these circuits.

282
00:26:49.798 --> 00:26:55.709
To tell medi, Cal them with the energy separation between any of that is something like Africa.

283
00:26:55.709 --> 00:26:59.878
And the environment that the circuit.

284
00:26:59.878 --> 00:27:04.288
And only provide 10 me to tell them that means that if you prepared the system and it's.

285
00:27:04.288 --> 00:27:09.209
Ground state, it's quantum mechanical. It will stay there because there's nothing in the environment.

286
00:27:09.209 --> 00:27:16.138
To give it enough energy to promote it to their wants. And how can you prepare the system? And it's hard to make any.

287
00:27:16.138 --> 00:27:21.148
Just wait the system will the system as a finite 1 you can get started anywhere.

288
00:27:21.148 --> 00:27:27.959
And it will just relax the ground state since there's nothing that can promote it elsewhere. It will end up there and stay there.

289
00:27:27.959 --> 00:27:35.909
That's very good news. So if you put your cell phone, in addition, French refrigerator, it's in it's quantum mechanical proxy.

290
00:27:37.019 --> 00:27:40.798
Well, not quite. Okay, so that's the good news.

291
00:27:40.798 --> 00:27:47.308
What's the bad news? Bad news is that well, you would like to control this Cuba for this. We'll try to write laser on our circuit.

292
00:27:47.308 --> 00:27:50.578
I'm not quite the bright laser, but the secretary.

293
00:27:50.578 --> 00:27:55.199
Which will be shy and migrate migrate source.

294
00:27:55.199 --> 00:28:02.368
Physically offensive and and try micro voltage. See a thing at micro frequencies.

295
00:28:02.368 --> 00:28:09.028
On this on this circuit is exactly like a laser. If you understand this.

296
00:28:09.028 --> 00:28:12.868
You should be able to send this also the same page.

297
00:28:12.868 --> 00:28:16.199
Um, and what you find is that it's.

298
00:28:16.199 --> 00:28:19.979
Since the energy separation is the same everywhere.

299
00:28:19.979 --> 00:28:24.058
Once this this voltage false as promoted the system.

300
00:28:24.058 --> 00:28:28.138
From 1 to 0T, there's nothing that prevented to go from 1 to do what to do.

301
00:28:28.138 --> 00:28:32.788
Trade and so forth. So at the end of the day, you end up with a superposition of.

302
00:28:32.788 --> 00:28:35.969
So called for you, and state of.

303
00:28:35.969 --> 00:28:41.278
The possible states in your Gilbert sticks, so you, there is major leakage. This is nothing humid.

304
00:28:41.278 --> 00:28:48.898
Okay, that's the problem. So you might remember that the why these.

305
00:28:48.898 --> 00:28:53.128
Atomic atoms were so good, was that the inner integration was not the same.

306
00:28:53.128 --> 00:28:56.638
And every for every, every okay.

307
00:28:56.638 --> 00:29:01.469
So, we need something which has not been here. We need a spectrum, which is not like this 1.

308
00:29:01.469 --> 00:29:05.128
So, we need a nothing new 11th and we have.

309
00:29:05.128 --> 00:29:09.088
1, such element, which is the junction.

310
00:29:09.088 --> 00:29:15.358
Well, this is not novel won the Nobel prize for this.

311
00:29:15.358 --> 00:29:18.959
Actually, something which discovered while he was a PhD student.

312
00:29:18.959 --> 00:29:21.989
All your guys be sparked a little bit harder.

313
00:29:21.989 --> 00:29:29.909
So so what I haven't told us is that if you fabricate the circuit.

314
00:29:29.909 --> 00:29:34.798
Which is basically a supervising wire.

315
00:29:34.798 --> 00:29:40.409
Then you, you put some outside, you put a layer of insulator on top of this wire.

316
00:29:40.409 --> 00:29:45.358
And then you glue us again wire. Okay. So basically you have a sandwich of Dr.

317
00:29:45.358 --> 00:29:48.479
Insulator Dr.

318
00:29:48.479 --> 00:29:55.078
Then current, we'll be able to go through this insulating there because of.

319
00:29:55.078 --> 00:30:01.348
Compliment effect, because of contract penalty, and there will be a specific relation between the current.

320
00:30:01.348 --> 00:30:04.378
That flows through this sandwich.

321
00:30:04.378 --> 00:30:08.098
And the flux integral of the voltage across the structure.

322
00:30:08.098 --> 00:30:11.098
That's the symbol for a junction.

323
00:30:11.098 --> 00:30:15.749
And the, what we're I'm writing here is that the current that flows through the sandwich.

324
00:30:15.749 --> 00:30:18.898
It's related to the flux, so the interval up the voltage.

325
00:30:18.898 --> 00:30:22.348
Actually, take all of the voltage across this.

326
00:30:22.348 --> 00:30:26.578
In this very special way. Okay. Why is this interesting?

327
00:30:26.578 --> 00:30:32.068
Let's look at this doctor here and you might know that the current is also related to the flux.

328
00:30:32.068 --> 00:30:36.898
Doctor, but with this very simple relation, there's a coefficient.

329
00:30:36.898 --> 00:30:41.338
Progress energy coefficient between threat and flux and that's what we call the inductance.

330
00:30:41.338 --> 00:30:45.269
Okay, now there is a relationship.

331
00:30:45.269 --> 00:30:49.858
But this relationship is not in here as is here, this is a nonlinear relationship.

332
00:30:49.858 --> 00:30:53.249
Actually, it's tempting to define this object.

333
00:30:53.249 --> 00:30:59.038
Okay, which if you again, this relation just get the L, you just get the documents.

334
00:30:59.038 --> 00:31:03.538
But now given this current Luxe relationship.

335
00:31:03.538 --> 00:31:09.058
You get this value for another call that Joseph sent and documents. So, what you find is that this sandwich.

336
00:31:09.058 --> 00:31:14.608
Yeah, it's just like an inductance. Well, not quite it's like an inductance, which relates to flux.

337
00:31:14.608 --> 00:31:17.818
But as a different relationship, and that relationship.

338
00:31:17.818 --> 00:31:22.288
Instead of being a simple parameter. There is some nonlinear function.

339
00:31:22.288 --> 00:31:27.989
Of the of the flux, so you might remember.

340
00:31:27.989 --> 00:31:31.528
That the energy associated to this, that.

341
00:31:31.528 --> 00:31:35.278
In the circuit was going as 1 over.

342
00:31:35.278 --> 00:31:38.429
Okay, so what is 1 of her out?

343
00:31:38.429 --> 00:31:43.919
That's Co, sign sign of 5. so that means that if I replace.

344
00:31:43.919 --> 00:31:49.588
In that transpire Jesus injection, but instead of a.

345
00:31:49.588 --> 00:31:54.028
Potential I get a CO signed potential. Okay.

346
00:31:54.028 --> 00:31:59.278
And I'm zooming in here, I'm focusing on 1 of these wells.

347
00:31:59.278 --> 00:32:03.118
And I really am according to basic quantum mechanics.

348
00:32:03.118 --> 00:32:07.288
Energy levels there, but because now I have something which is.

349
00:32:07.288 --> 00:32:11.939
but going to infinity but cutting off at some finite energy .

350
00:32:11.939 --> 00:32:16.919
They would have to re, adjust to that. And what you'll find is that.

351
00:32:16.919 --> 00:32:21.028
These levels are not uniform in the spread.

352
00:32:21.028 --> 00:32:24.568
Exactly what we want it, but this is our artificial.

353
00:32:24.568 --> 00:32:29.429
Okay, so I have this course potential which, which changed.

354
00:32:29.429 --> 00:32:33.808
The the spread of the energy levels, and now I have these.

355
00:32:33.808 --> 00:32:39.538
We've agreed, I still need to give you numbers of our. They are that's coming in a 2nd.

356
00:32:39.538 --> 00:32:43.318
Which will, we will be using as our.

357
00:32:43.318 --> 00:32:49.078
As I keep it states, so exactly. The question of our good is.

358
00:32:49.078 --> 00:32:56.098
Uh, the system, so 1st, how much time does it take using this voltage password to source?

359
00:32:56.098 --> 00:32:59.848
To create any Super position, for example, to go from 0T to 1.

360
00:32:59.848 --> 00:33:06.298
It's a few seconds with ions. We're, it's typically a few microseconds.

361
00:33:06.298 --> 00:33:12.058
And that's basically because the type of moment of these artificial atoms is larger.

362
00:33:12.058 --> 00:33:15.598
Just because there are larger objects, and they call a couple of more strongly.

363
00:33:15.598 --> 00:33:20.009
Did you feel so daytime is faster.

364
00:33:21.058 --> 00:33:30.179
The relaxation time, and a T, to time is not as good. And that's also because in a sense, the are also a couple more strongly to things. You didn't want.

365
00:33:30.179 --> 00:33:34.348
So this is a very rough plus of 2.

366
00:33:34.348 --> 00:33:42.358
In blue and team 1 in great as a function of year and I just did a few representative experiments.

367
00:33:42.358 --> 00:33:47.669
So, what I find for is very encouraging is that it's much better to see this on the log scale.

368
00:33:47.669 --> 00:33:51.449
Which tells you that yes, there is really improvement over the years.

369
00:33:51.449 --> 00:33:58.169
And what you find is that in the last 10 years, or so we've.

370
00:33:58.169 --> 00:34:02.249
Once when from few fraction of analysis again.

371
00:34:02.249 --> 00:34:06.419
2 teachers of the order now of 100 micro 2nd.

372
00:34:06.419 --> 00:34:10.349
So actually will be talking in many seconds, which.

373
00:34:10.349 --> 00:34:14.849
Very good news so point 3rd is much better than 100 microseconds.

374
00:34:14.849 --> 00:34:19.318
And so there is city improvement and I'll come back to.

375
00:34:19.318 --> 00:34:24.088
Uh, to this by the end of the talk, how these last points were.

376
00:34:24.088 --> 00:34:37.889
But this okay, so this was a good introduction. So if you like the topic here, you can now watch the rest of the video on your own, get more.

377
00:34:37.889 --> 00:34:48.358
Get more of the hardware here. If you want still more depth, there's this talk here.

378
00:34:48.358 --> 00:34:52.318
And by the way you notice, I'm selecting these talks.

379
00:34:52.318 --> 00:34:58.889
Some workshops and summer's goals and so on. So, for example, here you are free to go to the.

380
00:34:58.889 --> 00:35:07.559
Summer school, quantum computing, summer school here and find more talks and go to other years in 2002. okay. Here.

381
00:35:07.559 --> 00:35:12.418
You can read about it. I'm not going to start that.

382
00:35:12.418 --> 00:35:18.478
The same thing, I'm going to show 10 minutes of it to give you a feeling of it.

383
00:35:18.478 --> 00:35:21.599
Start around 6.

384
00:35:21.599 --> 00:35:24.599
Silence.

385
00:35:24.599 --> 00:35:28.349
Good.

386
00:35:28.349 --> 00:35:31.378
Silence.

387
00:35:31.378 --> 00:35:35.668
Our own hardness where every particle represented a bit.

388
00:35:35.668 --> 00:35:42.418
Then you could not store as much information as you could handle with 300 few events.

389
00:35:42.418 --> 00:35:51.028
That's amazing, isn't it? So.

390
00:35:51.028 --> 00:35:55.528
In October, like, just 2 months ago.

391
00:35:55.528 --> 00:36:01.048
Google came out and announced that they had reached something called quantum supremacy.

392
00:36:01.048 --> 00:36:07.438
How many of you have heard about that? Okay by quite many of you.

393
00:36:07.438 --> 00:36:14.548
Good so, in there, so they made it a quantum computer out of 53 Cubans.

394
00:36:15.719 --> 00:36:21.119
And they use superconducting. So what I'm going to tell you now, is what are those Cubans.

395
00:36:21.119 --> 00:36:25.139
That were in the Google quantum processor and how do they work.

396
00:36:27.659 --> 00:36:31.108
This.

397
00:36:31.108 --> 00:36:36.208
Yeah, we should switch to to.

398
00:36:36.208 --> 00:36:41.938
Okay, so here you see a superconducting tube and.

399
00:36:43.318 --> 00:36:51.239
The light grey here is aluminum. The dark gray is the Silicon on which this circuit the sitting.

400
00:36:51.239 --> 00:36:55.858
So this is the superconducting circuit accomplish Super productivity in a minute.

401
00:36:55.858 --> 00:36:59.398
And you see that this is essentially.

402
00:36:59.398 --> 00:37:04.469
Capacitance, and then there's a small component up here.

403
00:37:04.469 --> 00:37:08.278
That does all the trick, the whole circuit.

404
00:37:08.278 --> 00:37:15.958
Is like half a millimeter across, but this little circuit that does the trick is just 100 by 100 nanometer.

405
00:37:17.039 --> 00:37:21.449
And this can now act like an artificial after.

406
00:37:22.889 --> 00:37:25.889
And in particular, it can have.

407
00:37:25.889 --> 00:37:30.628
It it behaves quantum mechanical and it can show superposition.

408
00:37:30.628 --> 00:37:35.248
This was not obvious 30 years ago that that.

409
00:37:35.248 --> 00:37:38.579
Circuits can behave mechanically. They're so big.

410
00:37:38.579 --> 00:37:42.179
But still they can behave, sir quantum, mechanically.

411
00:37:44.219 --> 00:37:47.938
So, a few words about Super collectivity.

412
00:37:47.938 --> 00:37:52.199
So, Super connectivity is something that happens in metals.

413
00:37:52.199 --> 00:37:58.289
When you cooled down a metal there is and measure that the resistance of it.

414
00:37:58.289 --> 00:38:01.559
You see that at some point, at some temperature.

415
00:38:01.559 --> 00:38:05.159
The resistance drops to 0T. That's Super collectivity.

416
00:38:05.159 --> 00:38:11.070
There's also, so this is the most notable.

417
00:38:11.070 --> 00:38:15.630
Property of a Super conductor there's also another property.

418
00:38:15.630 --> 00:38:19.739
Which is that I did an expense magnetic fields.

419
00:38:19.739 --> 00:38:22.920
And that allows you to.

420
00:38:22.920 --> 00:38:25.949
And magnets on top of the Super conductor.

421
00:38:25.949 --> 00:38:30.030
So, it actually can leverage the.

422
00:38:30.030 --> 00:38:33.510
And you can see here that it just floats and in.

423
00:38:33.510 --> 00:38:37.320
Thanks.

424
00:38:38.940 --> 00:38:42.929
You may think that a Super conductivity is a rare.

425
00:38:42.929 --> 00:38:47.909
Phenomena, but actually, if you look at the periodic table.

426
00:38:47.909 --> 00:38:53.130
The blue elements you see here are all become superconducting at some temperature.

427
00:38:53.130 --> 00:38:57.840
The green ones become superconducting if you apply a pressure.

428
00:38:57.840 --> 00:39:01.199
But, in fact, Super conductivity is a very.

429
00:39:01.199 --> 00:39:04.440
You show phenomenon if you're at low enough temperatures.

430
00:39:06.510 --> 00:39:10.320
So, now, how do we, how do we put Super conductors into.

431
00:39:10.320 --> 00:39:15.840
How will we do? We make them into Cubans so 1st, I want to discuss.

432
00:39:15.840 --> 00:39:20.670
You know, the basic elements that you know about from, from electricity.

433
00:39:20.670 --> 00:39:24.389
The capacitor.

434
00:39:24.389 --> 00:39:30.780
So, I remind you that the current in the capacity is given by the capacitance time to the.

435
00:39:30.780 --> 00:39:34.650
Derivative of the voltage and conversely.

436
00:39:34.650 --> 00:39:38.489
In an doctor, the voltage is given by the.

437
00:39:38.489 --> 00:39:41.789
The inductance times the derivative of the current.

438
00:39:41.789 --> 00:39:49.800
Now, and so you make these type of components out of Super conductors and that means you have no loss.

439
00:39:49.800 --> 00:39:54.000
So, if you lose energy, you also lose information.

440
00:39:54.000 --> 00:39:58.199
So this is an important part that you want to have something, which is lossless.

441
00:39:59.550 --> 00:40:03.929
So, to build an actual superconducting Cuba.

442
00:40:03.929 --> 00:40:09.300
You start out by just putting these together here in a residence circuit.

443
00:40:09.300 --> 00:40:13.679
So this is a resident circuit that will isolate that some frequency.

444
00:40:13.679 --> 00:40:21.059
And so charge will be on the capacitor, and it will go through the doctor go here. And then it'll go back again.

445
00:40:21.059 --> 00:40:26.909
And this will repeat itself many, many times. So typical frequencies are 5 gigahertz.

446
00:40:26.909 --> 00:40:31.320
The same frequencies that you have in your cell phones and WI Fi.

447
00:40:31.320 --> 00:40:35.250
But now, if you have a circuit like this.

448
00:40:35.250 --> 00:40:44.880
The energy spectrum is equitable, it means that the, if you supply energy and excite from 0T, then this state 0T to 1 here.

449
00:40:44.880 --> 00:40:50.010
You can then continue with the same energy to go to 2 and 2, 3 and so on.

450
00:40:50.010 --> 00:40:57.420
This is what's called the harmonic oscillator, but now, if we want to make this like an Atom.

451
00:40:57.420 --> 00:41:01.829
We, we need a non linearity.

452
00:41:01.829 --> 00:41:05.760
And this is where this component comes in.

453
00:41:05.760 --> 00:41:09.000
This here is called a Joseph's conjunction.

454
00:41:09.000 --> 00:41:13.349
This is a non linear element that acts like an indicator.

455
00:41:14.639 --> 00:41:19.619
But it's so nonlinear so that when you put in a single excitation.

456
00:41:19.619 --> 00:41:23.130
The quantum excitation into this.

457
00:41:24.659 --> 00:41:28.739
Just that 1st, automation or 1st.

458
00:41:28.739 --> 00:41:33.659
Changes the value of the inductance. So then when you.

459
00:41:33.659 --> 00:41:40.079
And that changes the residence frequency of this so that the next quantity you want to put in, you have to put in.

460
00:41:40.079 --> 00:41:44.849
At the different frequency, and that's what you see here. So you get a non equity.

461
00:41:44.849 --> 00:41:47.849
Energy level spacing.

462
00:41:47.849 --> 00:41:52.320
And so now if you buy.

463
00:41:52.320 --> 00:41:56.579
Radiation to this with.

464
00:41:56.579 --> 00:42:02.219
A precise frequency that corresponds to this level splitting. You can go up to this, but then.

465
00:42:02.219 --> 00:42:05.519
You don't get further because then you, there's no.

466
00:42:05.519 --> 00:42:09.989
No levels that you can you can go to and so.

467
00:42:11.489 --> 00:42:15.510
If you if you concentrate on only the 2 lowest levels.

468
00:42:15.510 --> 00:42:20.699
Then we call this a Cuban if we also include the higher levels.

469
00:42:20.699 --> 00:42:24.210
This is precisely like an Atom.

470
00:42:24.210 --> 00:42:29.309
So, we call this an artificial Atom.

471
00:42:30.510 --> 00:42:36.570
So, I'm actually going to skip that and go to to this 1. so.

472
00:42:36.570 --> 00:42:42.929
In this, this person here is Brian. Joseph.

473
00:42:42.929 --> 00:42:46.230
Uh, when he was 26 years old.

474
00:42:46.230 --> 00:42:52.199
He figured out what happens in a system where you have a Super conductor on 1 side.

475
00:42:52.199 --> 00:42:57.000
And insulating barrier, and then a Super conductor on the other side.

476
00:42:57.000 --> 00:43:04.170
So, he derived these very famous equations or these Joseph's and relations.

477
00:43:04.170 --> 00:43:09.570
And tell you how current and voltage depend on the properties of the superconducting.

478
00:43:09.570 --> 00:43:12.719
And the interesting thing is that if you.

479
00:43:12.719 --> 00:43:16.260
If you combine these 2 expressions, you can see.

480
00:43:16.260 --> 00:43:19.949
That this element looks like an inductance.

481
00:43:19.949 --> 00:43:24.210
It actually acts like a non non linear inductance.

482
00:43:24.210 --> 00:43:29.639
And that is precisely what we needed for the, for the rest of the circuit I have before.

483
00:43:29.639 --> 00:43:34.110
So, the Joseph's on is.

484
00:43:34.110 --> 00:43:38.070
Approximately equal to inductance it's a nonlinear.

485
00:43:39.090 --> 00:43:43.860
And now you can do 1 more tricks, and that is to.

486
00:43:43.860 --> 00:43:47.159
Put 2 of these in parallel.

487
00:43:47.159 --> 00:43:53.190
And without going into details, what happens then, is that you can actually make it to.

488
00:43:53.190 --> 00:43:56.639
So, by putting magnetic flux in this loop here.

489
00:43:56.639 --> 00:44:02.099
You can actually tune the documents in there for you can tune the frequency of argument.

490
00:44:04.320 --> 00:44:10.139
Now, so here you can see an SCM picture of these Joseph junctions.

491
00:44:10.139 --> 00:44:14.730
And the line with here is about a 100 nanometers but how do you make that?

492
00:44:14.730 --> 00:44:19.469
Well, here are Chalmers many many labs around the world.

493
00:44:19.469 --> 00:44:22.710
We have a clean rooms where we have.

494
00:44:22.710 --> 00:44:26.639
These tools elect from bibliography machines.

495
00:44:26.639 --> 00:44:32.369
Which you can use to make these things, it's a bit like how people do.

496
00:44:32.369 --> 00:44:35.400
Computers you draw a pattern.

497
00:44:35.400 --> 00:44:39.360
And you transfer it to in this case metal to aluminum.

498
00:44:39.360 --> 00:44:44.309
And then you can, you can make you mention that 1.

499
00:44:46.260 --> 00:44:53.849
So now we're interested in power. So, yesterday it was also talked about how does light and matter interact then.

500
00:44:53.849 --> 00:44:58.980
And I think Thomas, I convinced you that this is important both for.

501
00:44:58.980 --> 00:45:04.289
Everything from the Internet to your eye light matter interaction is really important.

502
00:45:04.289 --> 00:45:09.840
And so normally, how do you how do you look at that? Well, you have normal atoms.

503
00:45:09.840 --> 00:45:14.280
And you have lasers and you shine on the on the atoms.

504
00:45:14.280 --> 00:45:20.849
And then you see what happens and so typically you use optical frequencies, which.

505
00:45:20.849 --> 00:45:23.940
Amounts so visible.

506
00:45:23.940 --> 00:45:27.269
So, which amounts to tend to the 15 hurts.

507
00:45:27.269 --> 00:45:30.480
Well, that's quite high frequencies, but.

508
00:45:30.480 --> 00:45:35.909
We are using much lower frequencies and much lower energies in general.

509
00:45:35.909 --> 00:45:41.760
So, we use microwave frequencies, which is more like 10 to the 10 hurts.

510
00:45:43.199 --> 00:45:47.130
And we can then build our atoms here.

511
00:45:47.130 --> 00:45:53.639
And we can put them in transmission lines. So this is a transmission line. It's essentially a squashed.

512
00:45:53.639 --> 00:46:01.380
And we put our atom in here, and then we can send in microwaves. We can see how much is reflected. We can see how much.

513
00:46:01.380 --> 00:46:07.469
Is transmitted, and there are a number of interesting things about the system that makes it.

514
00:46:07.469 --> 00:46:12.300
Advantageous compared to using real atoms.

515
00:46:12.300 --> 00:46:15.929
So 1 of the things is that we can really engineer the atoms.

516
00:46:15.929 --> 00:46:19.409
We can, we can tailor the parameters of the Atom.

517
00:46:19.409 --> 00:46:23.130
We don't need to take what nature gives us. We can do it ourselves.

518
00:46:23.130 --> 00:46:26.579
The other thing is that we can place this at them.

519
00:46:26.579 --> 00:46:30.539
Everywhere anywhere on the chip, and it also stays there.

520
00:46:30.539 --> 00:46:36.059
If you put another a normal Atom, it both slogged away and attach to something else or something.

521
00:46:36.059 --> 00:46:40.559
Something like that, so we can, we can put it wherever you want.

522
00:46:40.559 --> 00:46:44.039
Moreover, this coupling between this atom and there and.

523
00:46:44.039 --> 00:46:47.610
And the light.

524
00:46:47.610 --> 00:46:56.760
Okay, so you're free to watch the rest of that of course, if you like, but I hope I've given you a case of it.

525
00:46:59.099 --> 00:47:03.269
And again, just something a little more technical.

526
00:47:03.269 --> 00:47:06.480
And show you just to start of this thing here.

527
00:47:07.829 --> 00:47:17.219
Going silence.

528
00:47:17.219 --> 00:47:21.300
Hi, my name is Connell Murray. I work at the IBM.

529
00:47:21.300 --> 00:47:27.719
Digital Watson Research Center you'll please excuse the illusion. Heavy title.

530
00:47:27.719 --> 00:47:32.159
That I chose, but I did it to emphasize how remarkable.

531
00:47:32.159 --> 00:47:36.539
It is that we take something as unwieldy as quantum mechanics.

532
00:47:36.539 --> 00:47:40.559
And we've been able to harness that into a viable.

533
00:47:40.559 --> 00:47:46.650
Computing platform and that's thanks to the work over several decades. Now.

534
00:47:46.650 --> 00:47:51.119
Of looking at how to mitigate the various loss mechanisms.

535
00:47:51.119 --> 00:47:58.559
That we see that are present and so in this talk, I'll briefly describe some aspects of quantum computing.

536
00:47:58.559 --> 00:48:01.739
And focus on the superconducting.

537
00:48:01.739 --> 00:48:07.349
I'll discuss some limitations that we see in the current performance metrics.

538
00:48:07.349 --> 00:48:11.159
In particular looking at the various loss mechanisms at play.

539
00:48:11.755 --> 00:48:23.425
I'll discuss how we can actually predict a certain aspect of electric loss within these systems. We can do that through a combination of either find an element or analytical approaches.

540
00:48:23.755 --> 00:48:26.755
And then I'll close with some discussions about.

541
00:48:27.059 --> 00:48:33.269
Um, the prospects that we expect to see from this near term quantum computing machine.

542
00:48:33.269 --> 00:48:36.630
So, if you look at the history of classical computing.

543
00:48:36.630 --> 00:48:40.409
It's been also pretty remarkable. That's starting.

544
00:48:40.409 --> 00:48:49.590
Up in the top, right there with 1 of the 1st integrated circuits about 50 years ago, where the device sizes were roughly 10 in size.

545
00:48:49.590 --> 00:48:52.829
We've been able now to get down to the bottom image.

546
00:48:52.829 --> 00:48:58.800
Of the case where we have things that are less than 10 nanometers for device.

547
00:48:58.800 --> 00:49:02.639
Advice lengths and again that's thanks to the ingenuity.

548
00:49:02.639 --> 00:49:08.130
And the resourcefulness of many scientists and engineers in being able to achieve that.

549
00:49:08.130 --> 00:49:14.429
That's been done basically through live graphic scaling. The problem is the past decade or so.

550
00:49:14.429 --> 00:49:19.320
That's been insufficient to also give us the performance gains that we require.

551
00:49:19.320 --> 00:49:23.309
Part of that is simply the nature of the materials with which we work.

552
00:49:23.309 --> 00:49:29.820
The fact that the sides they're used to insulate between the gate and the channel regions.

553
00:49:29.820 --> 00:49:39.840
Are so thin they're no longer insulating other cases where we actually have to tune the strength within the device channels to improve the carrier mobility.

554
00:49:39.840 --> 00:49:45.030
And then, in some cases, 2, we've actually moved from a planner geometry to 1 of 3 dimensions.

555
00:49:45.030 --> 00:49:50.849
That's being done so that we get better electrostatic control of the regions of interest.

556
00:49:50.849 --> 00:49:54.329
But even being able to accomplish all these.

557
00:49:54.329 --> 00:49:58.500
These feats, the problem is, we're hitting a thermodynamic hard limit now.

558
00:49:58.500 --> 00:50:09.059
And that's referred to as Bolton tyranny, the fact that there is a fundamental amount of voltage that has to be supply to these systems. So that we can distinguish between the on and off.

559
00:50:09.059 --> 00:50:18.210
And therefore, we have to look at different types of paradigms because the fact that these power densities are climbing dramatically.

560
00:50:18.210 --> 00:50:22.019
And I would argue that quantum computing is the leading contender.

561
00:50:22.019 --> 00:50:29.280
In that rates, so I'm showing a picture here that many of you have seen in various forms and how we describe.

562
00:50:29.280 --> 00:50:33.030
The difference between classical and quantum computing.

563
00:50:33.030 --> 00:50:41.579
The idea is that in the left, where we have classical bits that can either be a 1 or 2 States, we can either have current flowing corresponding to the 1 state.

564
00:50:41.579 --> 00:50:48.900
Or we have an off state in the picture on the right here, which is known as block spear.

565
00:50:48.900 --> 00:50:54.059
We have those 2 States as well. We have the ground state 0T state.

566
00:50:54.059 --> 00:51:02.010
Okay, so if you like this and you're going to get deeper later, you feel free to watch it. Let me see if I can get to the end of it and talk.

567
00:51:02.010 --> 00:51:05.699
You can talk about what is going to happen in the future.

568
00:51:05.699 --> 00:51:09.539
So, if we actually invert this picture, just.

569
00:51:12.090 --> 00:51:16.289
With the gates and the errors that are present within.

570
00:51:16.289 --> 00:51:20.760
Those 2 bits, particularly in this era of near term quantum.

571
00:51:20.760 --> 00:51:29.909
And so all of those aspects, although they're not necessarily dimensions. They do help to to define a parameter.

572
00:51:29.909 --> 00:51:34.530
That is hardware agnostic. It can actually tell you.

573
00:51:34.530 --> 00:51:42.000
What perhaps the most pivotal aspects of and 1 example I'm showing here on the bottom right? The fact that.

574
00:51:42.000 --> 00:51:45.000
Just certain past a certain threshold.

575
00:51:45.000 --> 00:51:52.440
You can add additional Cubans, but if the error rate isn't sufficient to have them be part of the entangling process.

576
00:51:52.440 --> 00:52:04.920
You will not get a change in the quantum volume that's present. Now. Thankfully, the quantum volume has increased dramatically over the past 3 years or so.

577
00:52:04.920 --> 00:52:14.369
There are a variety of different layouts that IBM Q offers and these are publicly accessible. Anyone can go online and be able to.

578
00:52:14.369 --> 00:52:23.429
Utilize these, and they have various differences with respect to the number of Cubics of course. But also how they're related, how they're interconnected.

579
00:52:23.429 --> 00:52:28.170
And ideally, you would want the greatest amount of interconnectivity present.

580
00:52:28.170 --> 00:52:32.849
But, not as a detriment of the error that might be introduced by having.

581
00:52:32.849 --> 00:52:36.869
Um, so many series with these systems.

582
00:52:36.869 --> 00:52:43.440
The point is that we have been able to demonstrate an exponential trend in the quantum volume.

583
00:52:43.440 --> 00:52:46.949
And, in fact, just earlier this year.

584
00:52:46.949 --> 00:52:50.280
We've announced the Raleigh generation.

585
00:52:50.280 --> 00:52:55.829
Of Cuban 28 cubic devices that have demonstrated volume of 32.

586
00:52:57.659 --> 00:53:06.869
And so why can we use these types of devices for clearly? There are examples that involve chemistry, material, science, machine, learning.

587
00:53:06.869 --> 00:53:15.030
The problem again is that we don't yet have the ability to perform true error correction within our Cubans that would represent.

588
00:53:15.030 --> 00:53:18.150
It fault tolerant, acute, quantum computing system.

589
00:53:18.150 --> 00:53:23.579
And instead we have ways in which we can try to mitigate that error loss. And I'll show that on the next slide.

590
00:53:23.579 --> 00:53:28.260
The fact that we do have universal computers mean that we can program these gates.

591
00:53:28.260 --> 00:53:34.409
And we try to look at what type of algorithms that have a short enough depth that can give us useful information from it.

592
00:53:34.409 --> 00:53:39.599
I also want to mention though, that represents an open source.

593
00:53:39.599 --> 00:53:46.530
Co, quantum computing software framework that's being widely adopted. Now over 300000 downloads they are present.

594
00:53:46.530 --> 00:53:49.889
In which you can also use to help program.

595
00:53:49.889 --> 00:53:54.539
Both simulations as well as actual experimental cases on these various.

596
00:53:54.539 --> 00:53:59.159
Devices and the 1 example here Hong Sean error medication.

597
00:53:59.159 --> 00:54:04.590
Involves the calculation of the ground state energy of the lithium hydride molecule.

598
00:54:04.590 --> 00:54:09.059
As a function of the distance between the 2 and the atoms.

599
00:54:09.059 --> 00:54:16.170
And, of course, you can see on the left that if we only use circuit depth of 1 here, we get a trend that.

600
00:54:16.170 --> 00:54:19.170
That does follow what we would expect.

601
00:54:19.170 --> 00:54:25.170
But there is a significant decrease in the accuracy that 1 would hope for.

602
00:54:25.170 --> 00:54:31.889
What I'm showing in the right then is by using a technique called Richardson extrapolation.

603
00:54:31.889 --> 00:54:37.860
You can actually see the performance based on a certain gate time.

604
00:54:37.860 --> 00:54:46.679
You can amplify that time and then actually use that to back propagate what you would expect to have in the condition. It's error.

605
00:54:46.679 --> 00:54:55.289
And that corresponds to the bottom plot here where now you actually can utilize the fact of a greater circuit them to get you more accurate information.

606
00:54:56.639 --> 00:55:00.030
And there are plenty of different examples of this.

607
00:55:00.030 --> 00:55:04.860
I would encourage you to access the IBM quantum Web site to see.

608
00:55:04.860 --> 00:55:09.239
How, in fact, some of these can be applied to particular research interests.

609
00:55:09.239 --> 00:55:12.869
That you might have, and with that, I'll conclude.

610
00:55:12.869 --> 00:55:18.480
We've gone over different ways in which we've been able to see how materials design processing are important.

611
00:55:18.480 --> 00:55:26.219
So, the operation of we're always looking for the next limiting mechanism that's present in our system.

612
00:55:26.219 --> 00:55:32.250
But in the meantime, the fact that we do have near term quantum computing applications that we can utilize.

613
00:55:32.250 --> 00:55:36.599
And they involve the fact that we are able to increase our quantum volume.

614
00:55:36.599 --> 00:55:44.190
And eventually, hopefully move us from the state of our mitigation to true error correction. And that will represent a giant leap.

615
00:55:44.190 --> 00:55:48.360
In terms of the ability to operate our quantum computers.

616
00:55:48.360 --> 00:55:52.019
Please access this website if you'd like more information. I'd like to. Thank.

617
00:55:52.019 --> 00:55:56.010
Um, the contributions from my colleagues here, and the hard work that.

618
00:55:56.010 --> 00:56:02.730
The entire IBM queue team does Thank you.

619
00:56:03.054 --> 00:56:17.664
Okay, give you a feeling for that. Q tech Academy. I just mentioned I've showed you some videos from it, and it's got various other things here. So feel free to.

620
00:56:18.210 --> 00:56:24.599
Browse that if you'd like more information or, like, ideas for a final project also to be talking about at some point.

621
00:56:24.599 --> 00:56:36.900
Okay, so we saw the bits. Okay different architecture is D, wave's architecture. So I've got some videos here some short ones and then we'll show.

622
00:56:36.900 --> 00:56:41.699
I'd have a longer 1 about how quantum kneeling works.

623
00:56:41.699 --> 00:56:52.289
And then following the videos, I'll point you at some slide to which side the same information on slides. It's fun to see the videos 1st, to give you a feeling of it.

624
00:56:52.289 --> 00:56:58.530
And this will lead in to continue this next week but let me start things off for you.

625
00:57:05.460 --> 00:57:09.420
Quantum computing, although being a relatively young field.

626
00:57:09.420 --> 00:57:15.690
It's actually quite complex, and there's many different approaches being pursued around the world for building a quantum computer.

627
00:57:15.690 --> 00:57:18.809
The D wave, our approach is quantum.

628
00:57:18.809 --> 00:57:24.389
And in this video, I'll explain what that is and how it relates to the other forms of quantum computing.

629
00:57:24.389 --> 00:57:30.030
So, the continent healing is basically a way of using.

630
00:57:30.030 --> 00:57:35.280
The intrinsic affects of quantum physics to help you solve certain types of problems.

631
00:57:35.280 --> 00:57:40.349
Called optimization problems and also a related program called probabilistic sampling.

632
00:57:41.394 --> 00:57:55.824
Now, let me explain what these are. So, an optimization problem is a problem where you try to search for the best configuration out of many, many different, possible combinations. So an example of this is say, you're trying to build a house.

633
00:57:55.885 --> 00:58:02.724
You've got a fixed budget to spend the many, many different things you would like to have in your house. Maybe you can't afford them all.

634
00:58:03.264 --> 00:58:11.965
So, the challenge is to find out what combination of all of those different things, you can afford to fit into your house and maximize your happiness.

635
00:58:12.324 --> 00:58:23.545
So, you can imagine spending your entire budget on house, which is great, or you spend your entire budget and is not quite so good. So, and optimization problem is trying to find the best configuration.

636
00:58:23.820 --> 00:58:34.199
The reason you can use physics to solve optimization problems is because you can frame them as a type form called an energy minimization problem.

637
00:58:34.199 --> 00:58:46.800
And, like, a fundamental part of physics is that everything is always trying to find this minimum energy state. So things slide down hills or thermodynamics, hop things.

638
00:58:46.800 --> 00:58:57.630
Cool down over time and it's also true of quantum physics. So quantum is using quantum physics to find the minimum energy state of something.

639
00:58:57.630 --> 00:59:01.230
So, solving problems.

640
00:59:01.704 --> 00:59:09.414
Related to optimization problems, but they're slightly different instead of focusing on trying to find the minimum energy state.

641
00:59:09.594 --> 00:59:18.264
What you're trying to do is sample from many low energy statements and try and characterize the shape of your energy landscape.

642
00:59:18.510 --> 00:59:21.780
This is useful for application errors like machine learning.

643
00:59:21.780 --> 00:59:27.659
Where you're trying to build a probabilistic representation of the world.

644
00:59:27.659 --> 00:59:36.210
And these samples give you information about what your model is like now, and you can use those to improve your models over time.

645
00:59:36.210 --> 00:59:45.150
Optimization problems also crop up in in machine learning and typically sampling problems and optimization problems are very difficult to solve and.

646
00:59:45.150 --> 00:59:51.480
On classical computers, so there's a lot of interest in trying to find alternative techniques to solving.

647
00:59:51.480 --> 01:00:01.889
These kinds of problems, so that's a description of quantum annealing and the kind of things it's used for.

648
01:00:01.889 --> 01:00:05.369
So, how does it relate to the other forms of quantum computing?

649
01:00:05.369 --> 01:00:10.320
Well, the 1st, 4 contribution that develops called gate model called.

650
01:00:10.320 --> 01:00:18.030
Animals and the differences between these 2 kinds, it can be summarize as follows. So.

651
01:00:18.030 --> 01:00:26.639
In quantum and healing what you're trying to do is harness the natural evolution of quantum states. Well, there you don't have any control over that evolution.

652
01:00:26.639 --> 01:00:30.119
So, you set up the problem at the beginning.

653
01:00:30.119 --> 01:00:37.619
Unit quantum physics do its natural evolution and the configuration at the end corresponds to the artsy or trying to.

654
01:00:37.619 --> 01:00:51.239
In game, on of quantum computing, the aim is a lot more ambitious. What you're trying to do there is be able to control and manipulate the evolution of that quantum state over time. Now, this is a lot more difficult because.

655
01:00:51.239 --> 01:00:54.480
Quantum systems tend to be incredibly delicate to work with.

656
01:00:54.480 --> 01:01:01.380
However, you can have having that amount of control means that you can solve a bigger class with problems.

657
01:01:01.380 --> 01:01:06.090
So, these differences, the reason why it's been impossible to scale up.

658
01:01:06.090 --> 01:01:10.170
Quantum annealing processes to over a 1000 cube. It's.

659
01:01:10.170 --> 01:01:15.809
Whereas the state of team game model, quantum computing is around 10 cube.

660
01:01:15.809 --> 01:01:22.409
So, it's a lot more difficult to get the cube bits to work together coherently in a gate model quantum computer.

661
01:01:22.409 --> 01:01:28.769
However, there have been some very powerful algorithms developed for use when they reach scale.

662
01:01:28.769 --> 01:01:42.389
So, a couple of examples are shows algorithm for factoring large numbers and also grove as algorithm for searching through databases. These promised to be way, way faster than anything. You could possibly run on a classical machine.

663
01:01:42.389 --> 01:01:55.289
Given our current knowledge there have been some other approaches to quantum computing have been shown to be equivalent to the gate model approach. And these are all known as universal quantum computers.

664
01:01:55.764 --> 01:02:08.304
So, to be classified as a universe, quantum computer, it needs to be showing that there's a mapping of the specialized gate model algorithms to these other forms that it doesn't take up too much time.

665
01:02:08.304 --> 01:02:12.594
So there's a polynomial time mapping and a polynomial resource mapping.

666
01:02:12.809 --> 01:02:15.809
From the game model approach to these other approaches.

667
01:02:16.829 --> 01:02:20.280
Isn't a universal quantum computer.

668
01:02:20.280 --> 01:02:24.030
However, it is related to 1 of the forms of.

669
01:02:24.030 --> 01:02:28.260
Universal quantum computers, a form called adiabatic, quantum computing.

670
01:02:28.260 --> 01:02:34.469
In fact, adiabatic quantum computing is a specific form of quantum annealing.

671
01:02:34.469 --> 01:02:38.730
Which also worked on the process of energy minimization.

672
01:02:38.730 --> 01:02:44.909
So, it's just to say that quantum annealing and universe called computers are completely separate entities. There is a.

673
01:02:44.909 --> 01:02:48.539
A link between, so.

674
01:02:48.539 --> 01:02:52.199
There are described quantum and healing and.

675
01:02:52.199 --> 01:02:56.760
The kind of things it's used for in my next video, I'm going to explain.

676
01:02:56.760 --> 01:03:04.710
Meeting works.

677
01:03:09.929 --> 01:03:19.409
Hello.

678
01:03:29.039 --> 01:03:40.230
My last video explain how relates to the other forms of.

679
01:03:40.230 --> 01:03:48.989
And also the kind of problems, continuity and so this video, I'm going to explain how the process that actually works.

680
01:03:50.309 --> 01:03:55.889
So, let's start by looking at a single cube and cube can be in a state of 0. 1.

681
01:03:55.889 --> 01:03:59.130
And these states from coated in a circulating current.

682
01:03:59.130 --> 01:04:08.699
A corresponding magnetic field, because if it's a quantum okay, you can be in a superposition of 0T state and the 1 state at the same time.

683
01:04:08.699 --> 01:04:12.360
And in the quantum healing process, what happens is.

684
01:04:12.360 --> 01:04:20.369
The goes from the superposition state into either the 0T state or the 1 state, which are classical States at the end of the end.

685
01:04:20.369 --> 01:04:24.360
The physics of this process can be shown in an energy diagram.

686
01:04:25.469 --> 01:04:36.210
To begin with, there's just 1 Valley, and the lowest point corresponds with the superposition state of the.

687
01:04:36.210 --> 01:04:41.940
Calls meetings are on a barrier is raised and this turns the energy diagram.

688
01:04:41.940 --> 01:04:52.980
Into what's known as a double well potential here the low point of the left Valley corresponds to the 0T state and the low point on the right Valley corresponds to the 1 state.

689
01:04:52.980 --> 01:05:00.719
And the Cuban will end up in 1 of these valleys at the end of the Neal. How does it decide which? 1? Well.

690
01:05:00.719 --> 01:05:06.659
Everything else being equal the probability of the cubic ending in the 0T to 1 the state is.

691
01:05:06.659 --> 01:05:10.289
Eva, it's for 50% triangle and any other state.

692
01:05:10.289 --> 01:05:16.530
And the interesting thing is, you can actually control the profitability of it falling into the 0T or 1 state.

693
01:05:16.530 --> 01:05:20.670
Which is done by applying an external magnetic field to the.

694
01:05:20.670 --> 01:05:23.969
And effectiveness is to tilt the double well potential.

695
01:05:23.969 --> 01:05:27.389
Increasing the probability of ending up in the lower. Well.

696
01:05:28.469 --> 01:05:31.769
This external magnetic field is called a bias.

697
01:05:31.769 --> 01:05:37.349
And the key was basically minimizing its energy in the presence of this external magnetic field.

698
01:05:43.885 --> 01:05:56.184
Okay, so being able to control the probability that the cube, but will fall into the 0T the 1 States is a really useful feature, but the real power of these processes comes when you start linking them together and they can start influence each other.

699
01:05:56.460 --> 01:06:04.530
This is done with a device called a capital and a couple basically defines how to influence each other.

700
01:06:11.070 --> 01:06:20.550
So, a couple can make the 2 couple Cubics 1 to end up in the same state. So that's either both 0T or both. 1.

701
01:06:20.550 --> 01:06:26.519
Conference and care, which 1 of these it is as long as the canvas end up with the same stage to each other.

702
01:06:26.519 --> 01:06:35.789
Alternatively, the coupler can make the neighboring cue that's want to be in the opposite States. So either 0, 1 or 1. 0.

703
01:06:36.960 --> 01:06:42.809
When you've got a coupling between 2 people that you're now using another phenomena in quantum physics called entanglement.

704
01:06:42.809 --> 01:06:48.119
When 2 are entangled, they now have to be considered as a single object.

705
01:06:48.119 --> 01:06:58.349
And now, which has got 4 States, so you can imagine a potential 4 States each 1 corresponding to a different combination of the 2.

706
01:06:58.349 --> 01:07:04.559
The relative energies of these states depend on the biases on each Cupid and the coupling between.

707
01:07:06.510 --> 01:07:17.909
Now, when I said that for couple wants the Cubans to be the same, or wants them to be, obviously what I really mean as the couples are making those states energetically favorable.

708
01:07:17.909 --> 01:07:25.650
So, if they're a couple of wants, the tube is to be the same. Really? What it's doing is lowering the energy of those 2 States in comparison to the other states.

709
01:07:25.650 --> 01:07:31.920
And if so, a couple of wants them to the opposite, then it's lowering the energy of those states.

710
01:07:31.920 --> 01:07:35.190
Team they can have a fire supply to it.

711
01:07:35.190 --> 01:07:42.989
Achievements can interact are the conflicts, and as a user, you can actually choose all the values for these biases and.

712
01:07:42.989 --> 01:07:46.019
Both the direction of them, and also the strength.

713
01:07:46.019 --> 01:07:55.380
This is basically hiring program. A quantum computer you choose a whole set of bias is a whole set of complains that defines an energy landscape.

714
01:07:55.380 --> 01:08:00.750
And then the quantum does content naming to solve or find the minimum energy of that.

715
01:08:03.599 --> 01:08:06.960
So, now you can start to see some of the complexity of these machines.

716
01:08:06.960 --> 01:08:12.059
A few bits I've got 4 States, I can define an energy landscape over.

717
01:08:12.059 --> 01:08:16.050
Up to 3 events, the number of States goes up to 8.

718
01:08:16.050 --> 01:08:27.810
And for each extra cube that I add, I actually double the number of states I can define this energy landscape over. So the number of States goes up exponentially with a number of few bits specifically.

719
01:08:27.810 --> 01:08:33.239
That relationship is 2 to the power of and where is the number of Cubics.

720
01:08:34.680 --> 01:08:41.609
Okay, so let me summarize what I've talked to.

721
01:08:41.609 --> 01:08:45.180
And content, meaning you start off with a large set of cube it.

722
01:08:45.180 --> 01:08:48.659
In each team is in a superposition state of 0, 1.

723
01:08:48.659 --> 01:08:58.979
And also, they're not connected yet, then they undergo the continent healing whether, and the biases get introduced and the cube is all become untangled.

724
01:08:58.979 --> 01:09:05.760
This large quantum object then changes the probability that each cube it will end up in the 0, 1 state.

725
01:09:05.760 --> 01:09:11.789
And then finally, at the end of the mail, each cube it ends up is either 0T or 1.

726
01:09:11.789 --> 01:09:17.250
Was final state is the minimum energy state of your problem, or 1 very close to it.

727
01:09:17.250 --> 01:09:21.149
And all of this happens in our chips and around 20 Micro.

728
01:09:31.800 --> 01:09:38.550
Okay.

729
01:09:38.550 --> 01:09:48.869
Still here, so that was a nice introduction to D wave and then we'll see a longer video part of all of it next time and feel free to watch ahead of me.

730
01:09:50.274 --> 01:10:05.185
You can watch this on the weekend if you'd like, and even they create yourself an account on the D wave as you're to start playing with it. If you want stuff written down, you can look at these slides here there in greater depth than the videos. So far.

731
01:10:05.460 --> 01:10:10.920
And also freedom to read ahead if you would like, it's not required, but you're certainly welcome to.

732
01:10:10.920 --> 01:10:18.449
Okay, so that's the material for today. We sell some IBM hardware. We saw some artificial Adams and so on introduction.

733
01:10:18.449 --> 01:10:30.989
To D, wave so mentioned to tech academy and so on. So have a good weekend and I will hang around to see if anyone has any questions.

734
01:10:30.989 --> 01:10:41.850
Other than that, look at the nice leaves. I'm looking at my window, this nice sleeves there and relax start work again in a day or 2. Thank you.

735
01:10:41.850 --> 01:10:51.090
Silence.

736
01:11:19.739 --> 01:11:23.760
No questions. Okay. Good. Bye guys.