WEBVTT 00:00.000 --> 00:13.320 Great, yeah, let's get going. Thanks, everyone, for coming to the Devroom, very exciting 00:13.320 --> 00:21.680 to see so many people around. Just want to give a shout out to the Devroom organizer, my 00:21.760 --> 00:29.600 myself and Nate from the Unitary Foundation, sort of like Linux Foundation, maybe more 00:29.600 --> 00:36.640 modest than that, but for quantum computing, Unitary because if you know a bit of 00:36.640 --> 00:43.440 linear algebra, yeah, Unitary matrices are important in quantum computing and that's 00:43.440 --> 00:49.720 all the linear algebra I'm going to use here. That's the email if you have a feedback 00:49.800 --> 00:57.320 for the organizer and Devroom manager. So a bit of history on the quantum computing Devroom, 00:57.320 --> 01:05.160 there were two addition, I think only two addition in the past, if I'm correct, 2019 and 2020, 01:06.200 --> 01:12.360 it was organized by the quantum open source foundation, it's another foundation, 01:12.680 --> 01:20.680 quantum computing specifically by Thomas Babay and Mark Fingerhood. And since then, 01:21.560 --> 01:30.840 there has been a four-year, had this bit for, you know, COVID and also I guess people have been 01:30.840 --> 01:36.600 very busy building quantum computers. They didn't have time to organize the Devroom, but I can 01:36.680 --> 01:42.840 guarantee there was lots of progress in the last four years. Yeah, and this is a bit of an 01:42.840 --> 01:49.320 inception, so this was a talk by Thomas, the organizer, this was the welcome in 2020, 01:50.120 --> 01:58.600 and he was resharing a tweet of me from 2019 showing how many people there was, so as popular as this 01:58.600 --> 02:05.800 one. Yeah, unfortunately, we missed the last four years, but we are back. I give a quick 02:05.800 --> 02:14.920 intro just to set the seniors, so I'll talk in one slides about quantum computing for people who 02:14.920 --> 02:22.360 come from like classical Devroom. I'll talk about why you need software for quantum computing, 02:22.840 --> 02:29.640 specifically why you need open source software for quantum computing, and why open source 02:30.280 --> 02:36.600 software for quantum computing is here at first them today. So quantum computing in one slide, 02:36.600 --> 02:43.480 this basically a paradigm that uses some property of quantum mechanics. They are called with these 02:43.480 --> 02:48.520 three names, you don't need to know all the details, but some of you are from quantum, so they already 02:48.760 --> 02:54.280 know superposition and tangleman. And interference, thanks to this awesome property of quantum 02:54.280 --> 03:00.840 mechanics, you can basically improve the speed and the security of computers for certain tasks. 03:01.880 --> 03:08.520 So there are many caveats in the last sentence, when I say improve the speed and security of computers, 03:08.520 --> 03:13.400 there are many caveats, but hopefully in the talks of today we will understand some of them. 03:14.040 --> 03:20.440 It's been an area of research for maybe more than 20 years, and we are approaching the 03:22.040 --> 03:28.040 the time where we are trying to make this like feasible, we are actually trying to build quantum 03:28.040 --> 03:35.160 computer and hopefully proper application and feasibility will come in the 20 years in the next 03:35.160 --> 03:40.520 20 years. I know that's one of the frequently asked questions when are we going to have application 03:40.680 --> 03:47.640 when to computer, so that's my answer at least. Right now, we are in this era of noisy 03:47.640 --> 03:55.160 intermediate scale quantum computer, which means we are around implementation that have up to 03:55.160 --> 04:02.920 1000 qubits. The application, since I know you were going to ask our like, 04:03.560 --> 04:11.960 you know, there is lots of hype on many, many application. I'm not even going to repeat those, 04:11.960 --> 04:17.480 but the actual application are cryptography, so you can actually compute the secret key from 04:18.040 --> 04:26.760 the public key of RSA scheme, so you can break some protocols for classical cryptography. 04:27.720 --> 04:33.640 Probably what's the most important is that you can simulate quantum systems, and this will allow 04:33.640 --> 04:39.880 you to design new materials, design new drugs, and apparently something that people really care about, 04:39.880 --> 04:46.040 design more efficient fertilizers. Lastly, there are some optimization problem that you can also 04:46.040 --> 04:52.520 speed up with quantum computer that are like things like routing and scheduling. 04:52.840 --> 05:03.160 But about software, so let's go through the job or the workflow, quantum scientists of 05:03.160 --> 05:09.800 a quantum engineer, so you come up with an algorithm or a protocol in this new paradigm, 05:10.600 --> 05:15.800 and you write it in a paper in like a research paper in a formal language. Usually, 05:15.800 --> 05:23.160 something like looks like this, so you can see these as something like a Boolean circuit, but more 05:23.160 --> 05:30.360 complicated than that, and then you actually want to run it on hardware, on quantum device. 05:30.360 --> 05:36.840 So there are many steps here, well first of all, when writing in a formal language means that 05:36.840 --> 05:42.120 you need to define the formal language, then you need to have something that compiles it, 05:42.280 --> 05:47.000 you want to test it first for at a small scale on like classical device, because you can do that, 05:47.000 --> 05:53.400 you can simulate it, and then you want to run it on hardware device. So that's what a quantum 05:53.400 --> 05:59.400 scientist or engineer would do. So I need some tools for this job, which are first of all, 05:59.400 --> 06:05.720 coming up with algorithm, usually some algorithm that we use in this noise intermediated 06:05.720 --> 06:14.120 are like hybrid algorithms, so you need some classical programming as well, things like error mitigation, 06:14.120 --> 06:18.680 BQA, which you don't need to know the specific, but basically you run a bit of quantum 06:18.680 --> 06:25.560 algorithm, a bit of classical algorithm, it and then you iterate on that, and I also say you 06:25.560 --> 06:30.760 need to write in a formal language, so what people are doing this day is defining some domain 06:30.840 --> 06:37.320 specific languages, this is probably what you are going to do for things like kiskit, 06:37.320 --> 06:45.320 silk, all these big companies, but also like I think we will see some of these languages defined, 06:45.960 --> 06:52.840 and you do have to define compilers, because of step 3, you need to define simulator, 06:52.840 --> 06:57.080 and these are all like things that you can write with classical software, like you can write 06:57.240 --> 07:03.240 these things in Python, then you need to control the hardware, so only the cylinder there that 07:03.240 --> 07:07.960 you see is the quantum stuff, all the rest is like firmware that you need to control will like 07:09.480 --> 07:18.280 classical toolkit, so how about open source then, well we are in a, it's a very interesting 07:18.280 --> 07:22.760 field for open source and a bit different from like other things that you see in other 07:22.760 --> 07:30.360 bedroom, because we are still in a R&D phase, so I personally believe open source is even more 07:30.360 --> 07:36.200 important here, because we don't know yet what the correct approach is, on top of that there are 07:36.200 --> 07:42.360 different approaches, that different research group, different companies are taking, so you do want to 07:42.360 --> 07:47.800 verify the claims they make, and you know if you have more transparency, if you have like open 07:47.800 --> 07:54.440 benchmark, this is going to be easier, finally we are here writing a full stack of a computer from 07:54.440 --> 08:00.280 like the hardware, like laser or things I don't understand, because I'm right, now the physicist 08:00.280 --> 08:04.680 but then you need to write like firmware on top of that, you need to write the compiler, 08:04.680 --> 08:10.440 you need to write like error correction, things that we wrote like 30, 40 years ago in like classical 08:10.440 --> 08:17.080 computing, and yeah we have the chance to redefine the whole stack, so we may also do it in open source 08:17.080 --> 08:27.000 way, incidentally like a week ago or so, Boba and Melco came up with this paper where they describe 08:27.000 --> 08:33.960 the economics of open source quantum computer, which is very interesting, the landscape is crazy, 08:33.960 --> 08:41.880 is there are like 400 project right now, only open source dedicated to quantum not only computation 08:41.960 --> 08:49.080 but device cryptography all these things, as a unitary foundation we do run a survey with 08:49.800 --> 08:56.840 every year with 800 participants, where we ask question about quantum open source, and yeah you can 08:56.840 --> 09:04.920 check it out at that link, and finally at first them, so this is the schedule of the day, 09:05.880 --> 09:15.080 some very interesting talk and yeah looking forward to them, and that was about it, yeah thank you. 09:23.720 --> 09:30.520 So I guess throughout the schedule there are like five minutes gap, we are going to take 09:30.600 --> 09:35.160 some of the questions at the end of the talk, and then maybe we'll run over a bit in this five 09:35.160 --> 09:40.520 minutes, but also use the five minutes to like leave the room if you decide that you don't really 09:40.520 --> 09:49.960 care about quantum computing, and let other people come in. So yeah if you have any question for me 09:50.520 --> 09:57.880 in time, both as an organizer and specifically to the thing I thought about very quickly. 10:10.120 --> 10:12.840 If you have an open source software, there is a build on top of what's sorry, 10:12.920 --> 10:25.640 proprietary hardware, I guess I don't know if it's question for me or for someone from the 10:26.600 --> 10:33.320 maybe some other open source policy debroom, but I don't know, I think that you know if it's on 10:34.760 --> 10:40.120 GitHub and it's public, and it defines one of the layers I was talking about, I would consider it 10:40.200 --> 10:54.600 open source, yes. Yes. Yeah that's a good question. So what the question is what will happen 10:54.600 --> 11:02.680 with our bank accounts, credit cards, that all rely on RSA cryptography? So I guess the answer is 11:02.680 --> 11:09.960 too full, there is like post quantum cryptography in like classical computing, so you can define 11:09.960 --> 11:17.320 protocol that are more complex than RSA. We don't yet know how well they perform, because you know 11:17.320 --> 11:24.760 RSA has been tested for like 30 years, and there are not like theoretical proofs, it's mostly 11:24.760 --> 11:29.160 based on the fact that we don't know how to solve this problem, like if you know RSA works, 11:29.720 --> 11:34.360 it's basically relies on this one-way function, and we don't know how to reverse them, so that's why 11:34.360 --> 11:39.800 we use them, but we also know that we tried to crack them for 30 years and we didn't succeed, 11:40.680 --> 11:47.080 so we can write protocols that are robust against quantum computing, that's one answer. 11:47.080 --> 11:55.000 The other answer, magically I said, magically quantum mechanics offers a way to do like 11:55.080 --> 12:02.680 cryptography, that some kind of cryptographic protocol that you can not do on like classical, 12:02.680 --> 12:08.840 with classical protocols, so maybe you can use that. And yeah, I think the times it's up, 12:10.920 --> 12:18.280 we will start in like one minute, and with the next talk.