Skip to main content

tv   Quantum Computing Research Applications  CSPAN  May 25, 2018 6:22pm-7:45pm EDT

6:22 pm
the u.s. olympic committee ceo testifies on capitol hill, and apologizes to athletes who were sexually abused. and on c-span 3, students impacted by gun violence share their experiences with the house democratic gun violence prevention task force. all of that tonight at 8:00 p.m. eastern. next a look at quantum computing. a computing technology that would operate faster and more efficiently than the computers used today. members of a house subcommittee focused their questions on the federal investment needed to insure the u.s. has a competitive advantage in the emerging field this is about 1:20. >> well, good morning and again,
6:23 pm
i'd like to welcome you all to the subcommittee on digital commerce and consumer protection here on energy and commerce. and as i mentioned, we have another subcommittee that's running now. and we'll have members coming back from the first floor upstairs during the committee from one to the other. i thank you all for being here today. i recognize myself for my five-minute opening statement. and again, welcome to the subcommittee and today's disruptor series examining quantum computing. we continue our disrupter service as we examine emerging tech zwris, supporting u.s. innovation and jobs. this morning we're discussing the revolutionary technology known as quantum computing. this involves harnessing the power of physics at its most basic level. unlike the computers we are familiar with that we use today, a quantum computer holds the potential to be faster and more
6:24 pm
powerful. this innovation is expected to change every industry and make problems that are impossible to solve today something that can be solved in a matter of days or weeks. efforts to develop a commercial a available and practical quantum computer are being pursued around the world. because of the costs involved in developing, a suitable environment for a quantum computer to operate, many efforts involve government support. both the european union and china have pledged to have already spent billions to develop a quantum computer. in the united states, development of quantum computer is proceeding at the academic, governmental and private sectors. in addition to the larger and familiar technology companies, smaller start-ups are leading efforts in this area. we're fortunate to have one of these start-ups, ion q to testify today. although a quantum computer holds a tremendous potential to solve previously noncomputable problems, there are skeptic who is question whether it will be possible to ever develop such
6:25 pm
technology. we look forward to our witnesses giving us their thoughts on this question. on the other hand some fear that the threats such a computer would pose to traditional computing model. especially when it comes to encryption and data security, some fear that a quantum computer would make it nearly impossible to keep future computers secure. data security and privacy are key concerns of this committee. we look forward to our witnesses addressing this issue as well. quantum computers hold tremendous potential to help solve problems involving discovery of new drugs, develop more efficient supply chains and logistics operations, searching massive bits of data and developing artificial intelligence. whichever nation first develops a practical quantum computer will have a tremendous advantage. how the united states is doing in that race. this is obviously a very dense subject. we are, we also understand there are several other areas, under development leveraging the
6:26 pm
principle of quantum mechanics, our goal today is simple, develop a better understanding of the potential of quantum computers, the obstacles to developing this technology and help spur innovation, competition and insure a strong and prepared workforce for future jobs, as we explore the topic, i would like to thank witnesses for coming to share their expertise on this complicated and revolutionary technology. i appreciate you being here today. at this time i will yield back my time and recognize the gentlelady from illinois, the ranking member of the subcommittee for five minutes. >> i want to thank you, mr. chairman. we continue our disruptor series with exploration of quantum computing. i want to congratulate all of you for being so smart. i was, dr. franklin and i was just told that i think your
6:27 pm
mother and i graduated from the university of illinois. about the same time. this was a time before we knew anything about computers, really. it was just beginning. and here you are today, the next generation leading us into the future. so i appreciate all of you being here today. this technology i understand is still in the research faphase, t the potential applications are tremendous, from health care to energy efficiency and everything in between. given this potential, global competitors from the european union to china are rushing to invest in quantum computing. the u.s. must make strategic investments if it wants to stay ahead. and those investments really start with s.t.e.m. education. we must encourage students, including young women, and students of color, to pursue interests in computer science and physics, fostering curiosity
6:28 pm
today. prepares young people, young minds to is become great innovators of tomorrow. as a former teacher, myself, i strongly believe that our future economic success depends on investing in our children's education. our research universities are leading the way on quantum computing. purpose investment is crucial. to develop technology until it can be profitably deployed in the private sector. however, the federal government has failed to provide robust, reliable investment in quantum computing. the lack of investment in s.t.e.m. education and research speaks to the misguided priorities of this republican congress. while wealthy shareholders get most of the gain from a $2 trillion republican tax bill, congress is underinvesting in students and research institutions. we fund tax cuts for the rich at the expense of our future
6:29 pm
prosperity. now that congress has passed a budget agreement, we have the chance to make some of the investments that our country so desperately needs. but instead of embracing the opportunity to advance bipartisan appropriations bills, the republican majority plans to bring up a recision bill to pull back funding for children's health insurance programs and other programs. and today, we will be voting on a bill to literally take food out of the mouths of families. we need to get our priorities straight. the u.s. can be a global leader in quantum computing and other groundbreaking technologies. but only if we prioritize investment for our future over tax cuts for the wealthy. i, i look forward to hearing from our panel. about the promise of quantum computing. i'll try my best to follow what you're telling me and the
6:30 pm
challenges that we face in developing this technology. i'm especially proud to welcome professor diana franklin from the university of chicago, the university of chicago is one of the leaders in quantum computing research. and i'm eager to hear more about this work so thank you chairman and i yield back. >> thank you very much. the gentlelady yields back. chairman the full committee has not made it in yet. is there anyone on the republican side wishing to claim this time? if not at this time that will conclude the member's opening statements and get to the real meet of the issue of the day that we want to tell but. i won't tell you how long ago, madam, ranker, how long when i took computer science in college. probably shouldn't say this. we used punch cards. and teletype machines. a bad saturday morning, went back to the computer science department and you were expecting about that much came back with that much and you knew you made a mistake.
6:31 pm
our witnesses will have an opportunity to make five-minute opening statements and our witnesses today are dr. matthew putnam, founder and ceo of a nanotronices. dr. christopher monroe, chief scientist and founder of ion q and professor of physics at the university of maryland. dr. diana franklin, professor and director of the computer science at the university of chicago, and mr. michael brett. ceo of q-brick. so again we appreciate your being here today and dr. putnam. are you recognized for five minutes for your opening statement if you would, just press that microphone and pull it close to you and we'll get started. thank you. >> thank you so much, chairman and congresswomen and congressmen. >> nanotronices does not make quantum computers, we the
6:32 pm
enabler of technologies and companies that with us strive to revolutionize the way information can be transformed. we've provided some of the world's largest companies and smaller entrepreneurial innovators with the tools of modern computation and imaging. we work with those that build the most advanced materials and microelectron microelectronics, nanotropics achieves this in the only way we see feasible for the continued exponential progression of technology. which is through artificially intelligent factories. quantum computing not only promises to break the barriers of encryption, it also breaks some fundamental barriers to human progress. many of our greatest achievements have been characterized in terms of competition and races. often, a technological race appears to be a war of ideologies or of business dominance. with quantum computing, there's an even greater battle. the fight against physical
6:33 pm
scarcity. there are three areas that we must work together on to win, not only for our nation, but for humanity. agriculture, new fertilizers can feed the increasing population of the world while maintaining diversity of crops. drug discovery by being able to simulate and produce molecules faster and with greater precision than are possible by traditional means. this will not only lead to cures for diseases, but reduce the often financially restrictive experimentation and trials that are required to make even incremental improvements in treatment. >> materials for power devices from batteries to solar cells. these have been studied for decades. but in many respects, the united states is still early on in this journey. companies are moving with speed and with national support, it's possible that quantum computing can soon reach an inflection point. the race to achieve a workable
6:34 pm
quantum computer, that can reduce scarcity to this level requires greater national attention than is currently been realized by either of the vast majority of companies or of the country as a whole. >> the steps to enabling quantum computing will need to involve one, an effort that funds the creation of factories for new quantum chips. a semiconductor fab for computers can cost as much as $20 billion. to a large extent, these fabs are not being built in the united states. we have an opportunity to acknowledge and to change this trend by leading the way in the construction of factories for this next generation of powerful computing. two, artificial intelligence. while quantum computing will increase the capabilities of artificial intelligence, the ability to design materials and
6:35 pm
software for quantum computers themselves, will come through the interaction of human and computer aejs. understanding such key elements as component design, fabrication conditions and the number of cubits needed requires collaboration of humans and machines. the number of cubits in a quantum computer is directly related to the number of calculations. a 10-cubit quantum computer can produce 1,000 calculations and a 30-cubit quantum computer can produce one billion. millions of cubits are required to achieve the full potential of quantum computing this exponential growth in cubit calculations is beyond the reach of factories as they are. without the advanced tools of ai for controlling factories, a truly useful quantum computer may not be possible. three, education. we need to develop the expertise required for the multidisciplinary nature of
6:36 pm
quantum computer science. it's physics, chemistry, mathematics, computer science, and application curiosity and expertise are all necessary. we cannot work in isolation. we need to embrace immigration and welcome strong talent from around the world. with expertise in these areas. >> when we look towards the future we can see it as a battle of ideologies of resources or of technologies. quantum computers encompass all of these to some extent. quantum mechanics is the basis of universal behavior at the smallest scales, but affects the largest of matter. it's therefore not surprising that harnessing the physical property has such far-reaching implications. because of this, it's important that we view it with the powerful associations that it warrants. with the weight of risk in a fractured world or of great rewards in a unified one.
6:37 pm
as we move forward, we see how quantum computing let's us scale in ways that meet not only the needs of industry, but of our country and the world. thank you very much. >> thank you for your testimony this morning. and dr. monroe, you're recognized for five minutes. thank you. >> thank you for the opportunity to testify, mr. chairman. i'm honored to be here for this committee's disruptor series on quantum computing. i'm a quantum physicist at the university of maryland and also co-founder and chief scientist at ion q, a start-up company that aims to build and manufacture small quantum computers. i've also worked with the national photonics initiative. which is collaborative alliance with industry and academics with the interest in developing quantum computing. we've promoted the idea of a
6:38 pm
national quantum initiative. there's pending legislation in the house science that's coming up in the house science committee. so i have about one minute to define what quantum computers are and i thinky get, get to some of the basics. we know that information is stored in bits. zeros or ones. the fundamental difference in quantum information is stored in quantum bits or q bits, these can be both 0 and 1 at the same time. as long as you don't look. at the end of the day you look and it randomly assumes one of the values. as long as you don't look there's a potential for massive parallelism. as you add q bits you get exponential storage capacities. and because quantum computers only work while you're not looking it involves quite revolutionary and even exotic hardware to realize this. individual atoms, that's the technology we use at ion q,
6:39 pm
superconducting circuits that are kept at very low temperatures, other competing platforms involve that type of technology. it's very exotic stuff. and i think -- i think within the next several years we are going to see small quantum computers with up to about 100 quantum bits. which sounds pretty small. even with 100 quantum bits it can in a sense deal with information that eclipses that of all the hard drives in the world. and on our way to a million q bits, where we can do new problems that conventional q bit computers could never tackle, we need to build the small ones first. so in terms of quantum applications, i would say it follows roughly into three categories, there are strong overlaps. in short term quantum sensors can enhance sensitivity to certain measurements that could impact navigation and it may be in a gps-blind environment and also remote sensing.
6:40 pm
the medium term, quantum communication networks may allow the transmission of information that can be provably secure because remember, quantum information only exists when nobody looks at it. if somebody looks at it, it changes. and that can make communication inherently secure. and in the long-term probably the most disruptive technology are quantum computers. and quantum computers are not just more powerful computers, they are radically different. they may allow us to solve problems that could never, ever be solved using classical computers, these involve optimization machines, economic and financial modeling and also the design of new materials and molecular function that could impact the health sciences and drug delivery, for instance. and even longer term, quantum computers could be used to do decorruption, breaking of popular codes. so there's a security aspect to everything, that quantum information touches. now, the challenges are, are
6:41 pm
pronounced in this field. there's, there are a few issues, one involving a workforce and one involving the marketplace. the workforce issue is that universities are chock full of students and faculty that are comfortable with quantum physics and we do research in the area, but we don't build things that can be used by somebody that doesn't want to or need to know all the details. where as industry makes those things, but they don't have a quantum engineering workforce, the marketplace is also a challenge. because we don't know exactly what the killer app for quantum computers will be. so we have promoted the idea of a national quantum initiative. that would establish several large and focused hub labs throughout the country. and other components as well. including user access program for existing quantum computers. it is imperative that the u.s. retain its leadership in this technological frontier, as we, as we heard from the chairman,
6:42 pm
there is, there is concerted efforts in europe and in particular, china, that is, that is spending lots of very focused investments in this field. so in conclusion, quantum technology is coming and the u.s. should lead in the next generation of sensors, computers and communication networks. the national quantum initiative provide as framework for implementing the quantum quantum initiative across the government. thank you for the opportunity to speak on quantum technology and the need for a nationally focused effort to advance quantum information science and technology in the u.s. >> thank you very much and dr. franklin, you're recognized for five minutes. >> thank you for the opportunity to testify, mr. chairman and ranking members. i'm honored to be before you to offer testimony on the importance of quantum technology, the big challenges we're facing in doing so my dual roles as director of computer science education at u chicago s.t.e.m. ed and a research
6:43 pm
associate professor in the department of computer science at the university of chicago i research technologies and computer science education. as lead investigator for quantum education for the epic quantum computing project in the nsf ex-editions in computing program. it's my job to develop a quantum computering workforce. quantum computing can be a game-changer in promising areas including drug design and food production. by accelerating research time, critical dollars could be saved along with improved quality of life. unlocking the secrets of nitrogen fixation could reduce the energy costs of fertilizer production and thus food production throughout the world. while the university has historically been on the forefront of computer science and emerging technologies, lapses in academic funding for quantum computer science have allowed global competitors to make great strides, putting the u.s. back ten years from where
6:44 pm
it could have been. in the past 17 years, since the inception of quantum computer science, distinguished from quantum physics and algorithm development, academic funding has only been available for eight of these years. leading to only 10 ph.d. students being trained, rather than a potential of almost 200 students and no meaningful education programs aimed at this area. as research groups came and went with the funding, post-docs were laid off and graduate students were transitioned to conventional computer science fields. yet universities are critical to commercialization. while companies work individually and compete against each other to produce proprietary tools, academics produce results in tools that all companies can improve on, as well as trained experts who can work at companies. they're both necessary for the commercialization of quantum computing. the challenge of bringing quantum dpurts to the point of usefulness cannot be overestimated.
6:45 pm
professor monroe knows extensive expertise in the former. i'm here to talk about the important role that computer scientists must take. funding in theoretical software and quantum devices has created a gap between the software which assumes large, perfect hardware and real hardware that is small and unreliable at this point. nsf has recently recognized this issue, supplementing their hardware initiative quantumly with a stack program that require as team. one gap is in software development. there's a difference between a quantum algorithm and software that can solve a particular problem. bridging this gap requires interdisciplinary team suches as exists in qx branch. deep expertise is necessary to figure out how to modify software that works in one specific context to another. much more so in quantum computing than in traditional computing. if this were furniture production, what we have now is piles of woods and screws, an expert needs to figure out how to use those to create furniture. what we want in the future is
6:46 pm
nor nonexperts to be able to go to quantum ikea, get a prefabbed kit and modify if for their application this exists for classical computing, but not for quantum computing. another gap is between software and hardware. current algorithms are written for perfect hardware. but hardware on the horizon is error-prone. we're on a journey to the perfect stuff, but not there yet. it's like if you planned to prepare a gourmet meal for ten, but when you arrive there's only supplies for six and you can only use the kitchen for two hours. would you need to adjust your plans, current and quantum computers that are on the horizon can only sustain computation for a limited time and they're very small. some modifications can be automated. however for more advanced modifications, the plan needs to be rethought. thus, some of the specific hardware limitations like the specific ways in which different technologies tend to introduce errors need to be communicated to the programmers so they can figure out how to adjust their
6:47 pm
applications. in order to realize quantum computing, federal funding needs to be first and foremost consistent. directed at not just building hardware and developing ail goe rihythms but to teams that are spread across a range of agencies with different missions, like nsf, daca dod and doe. there are more people available to write applications and to perform the engineering work at these companies. and above all, supporting the k 12 s.t.e.m. pipeline to train the next generation of innovators. with a significant investment in hardware, software and workforce development. i'm confident the united states can maintain its dominance in computing this concludes my remarks. i appreciate the opportunity to speak with subcommittee members and i'm happy to answer any questions you might have. >> thank you very much and mr. brit you're recognized for five minutes. >> thank you, chairman and ranking member and members of
6:48 pm
this committee. i'm thrilled to be here today to participate in today's hearing and discuss the opportunities and challenges presented by quantity up computing my name is michael brett, a ceo of a company called q branch, a data analytics company in washington, d.c. and with teams in australia and the uk. we're a fast-growing team with data scientists, software engineers and machine learning specialists who design ail goe rithic force challenging data problems, we're at the cutting edge of creating ail goe rithices and uncover other business insight that help our customers reduce their costs and serve their customers better. data analytics is already rapidly advancing technology area, delivering benefits to people all over the world but we're particularly excited about what quantum computing can do for our business. as we've heard, quantum computers are not just a faster computer, they enable an entirely different approach to performing calculations. in the realm of quantum physics, there's incredible and surprising phenomenon, that if harnessed could allow us to
6:49 pm
solve interesting and practically unsolvable problems, problems like simulating the interaction between molecules, as these molecules grow in size, the computational costs gross exponentially larger. our friends who build quantum computing hardware are in the process of creating machines that take advantage of this unique phenomena. these machines allow us as software developers to solve difficult problems using a different kind of mathematics, quantum math. much more efficiently than we ever could on classical computers. our ambition is simple, quantum computers will allow us to solve some of the most intractable and most valuable computational problems that exist today. these new solutions will benefit americans in ways they might not ever be aware of. globally the race is on to apply quantum computing problems to transport. energy production, health science and pharmacology, finance and insurance, defense and national security. we want our applications to be
6:50 pm
the first apps in a quantum app store. looking forward to the kind of quantum computers that are likely to be commercially available over the next decade, there are broadly three classes of application to become possiblein the near term. the first are optimization problems like logistics and transport routing. the second in machine learning accelerating computationally detecting patents and data sets. and using a quantum computer to simulate the behavior of molecules and materials and design new processes around them. across these three applications the potential value to every day citizens is immense. let me give you a concrete example. recently a study into quantum computing applications with merck, the pharmaceutical company. we designed a quantum algorithm and tested on today's available hardware to look at an approach
6:51 pm
at optimizing the production of a particular drug. the particular drug they are interested in has an extremely challenging production optimization process involved. quantum computing gave us the tools to lack at the manufacturing process in a different way that could radically change the efficiency of creating this drug and delivering value to the consumer. it's an application such as this that we are focused on a q branch. breakthroughs by a new approach in commuting that allows us to change the manufacturing process. there are challenges ahead in realizing this technology. the federal government can help us create the environment for industry to lead. the three challenges i would like to highlight, first, the skills in work force. as we have heard, if we are to be successful at bringing quantum computing to market we need a multidisciplinary work force with skills in quantum information science, a.i. combined with domain expertise
6:52 pm
in finance, pharmaceutical, and other universities. the second is an international cooperation. as american companies compete in this emerging ecosystem, we will achieve our fullest success through international corporation. there is valuable scientific research in engineering research made elsewhere including australia, the u.k., canada, japan and singapore. we need the best talent and technology globally. this means partnering. there will be national security considerations for this technology, of course, but restrictions are applied without due consideration it will stifle commercial innovation. finally, we need to maximize and leverage private sector investment. in the past 18 months we have seen an incredible acceleration in corporate r&d and venture capital. it's an exciting time. i must stress we are just at the beginning of this technology development. and the government can maximize
6:53 pm
and leverage this investment through targeted federal funding and coordination to reduce the gaps and overlaps in r&d and help accelerate the technology. i appreciate the opportunity to join you today and share with you about what we're doing at qx branch in quantum computing. this will give us a powerful new tool to create valuable software. >> thank you for your testimony. i appreciate all your testimony this morning. that will conclude our witnesses' testimony this morning. and we will begin our questioning from the members. and i will open with questions of five minutes. pardon my allergies this morning. it's that time of year in washington. first, i really appreciated reading your testimony last night, and a lot of questions in five minutes. but if i could start, dr. putnam, with you, if i may. i really was interested in what
6:54 pm
you said, what impact quantum computers would have in the united states. i have 65,000 manufacturing jobs and the largest farm income producing district in the state of ohio. on the manufacturing side, you talked about, with drugs and agriculture, energy, and this committee deals a lot with all of that, not really on the agricultural side. i was interested in that. and i would like to know, especially with the impact on manufacturing and also am i correct that it would create new opportunities while disrupting those existing industries out there today? >> thank you, chairman latta. my fellow ohioan. this is, of course, extremely personal to me as well being from ohio and being from, you know, creating and trying to enable manufacturing work. what's important, i think,
6:55 pm
about your question is that these are brand-new industries. it's not just about disrupting current industries. it's about creating jobs that are for the next generation of technology. and this is building, i think, interesting jobs as well for technologists of the future, and that goes through entire large factories. i mentioned the cost of a fab. it's not just the cost of building a fab. we'd like to bring down the cost to build fans. it's the opportunity for workers to be working with the latest of technologies. i think that the midwest and the rest of the country as a whole can only benefit from this. >> thank you. >> dr. monroe, what changes would be needed to ensure america has that work force that is ready for quantum computing revolution we have been hearing from the witnesses? we have to have the work force out there and the training. how do we get to that point? on the educational side, especially at the university
6:56 pm
levels, we need a university specialized in that in the field? what do we need to do? >> well, thank you for the question, chairman latta. there are a number of things that we can do as a country to foster this gap, this connection between university and government laboratory research and, as i said, industrial production. at the university side, i'm sorry to say that most engineering and computer science departments haven't really embraced this field as dr. franklin mentioned. >> why not? >> well, i have my own thoughts on that. i think actually my daughter is a computer science major at the university of maryland. the computer science department, the students are keen to get a high-paying job right after they graduate. quantum computing, not that it's not a high-paying job, but it's a speculative field, and it's hard to identify exactly what the marketplace is, and i
6:57 pm
think computer science departments and engineering departments, i think they have not embraced this field as much as the sciences have, and i think that is changing at some places. my university at university of maryland is one of those. chicago is another. several across the country have done that, but it's not widespread. many of these departments won't hire faculty that are doing research in this field. i think dr. franklin mentioned the national science foundation is taking an active role in trying to change that by instituting new grant programs that foster the development of quantum computer science. on the industry side, it's a tough nut to crack because this new technology, as i mentioned, involves very exotic type hardware that industry doesn't have so much experience with. it reminds me of, in history in the '50s, when devices were
6:58 pm
being developed and scaled. the people who did this over the many decades, including gordon moore who founded intel, these were not vacuum tube engineers who instituted the previous generation of computers. so it takes time and it takes risk and it takes funding from these corporations to do that. >> thank you very much. and my time is about to expire. i will yield back and recognize the member from illinois. >> i'm starting to understand taking a quantum leap because really what you're talking about is, of all the things we have heard about, the most disruptive in a good way and a challenging way to the future. and so i wanted to talk to dr. franklin. i think i know about education.
6:59 pm
and i do want to hear about, more about epic and the things that you are doing. but first i want to hear about your efforts with younger students in a minute. i want to first hear about what's happening at the graduate and undergraduate level. you know, what i'm hearing from all of you is that work force capacity is really a challenging issue. and if we are going to be competitive and if we are going to keep up with countries that are making the eu and also china, then we need to get serious about making these public investments. i'm wondering if you can talk to me a little bit about the urgent need. >> yes. so i think dr. monroe mentioned that computer science hasn't had as much quantum in it. i think it comes back to the funding lapses. our group and other groups started, and the way courses get created is that graduate
7:00 pm
students get trained in a field. they go out and become professors, create classes and train more students. those students need to be able to have jobs in order to make it worth it for them to take those courses. if no federal funding, if a program gets canceled and you are two of six and all of the federal funding goes away, and the graduate students get put in other fields, you are not going to have an education program. so that's what's happened twice, is that the federal funding went completely away for the computer science portion of quantum computing. so groups that were active and getting into the field left the field. and so now, with this new stack funding and the new epic program that we have, we are planning educational initiatives at all levels, including tutorials for professionals. we have tutorial in june and tutorial in october for professors and graduate students who are already in the field who want to transition to quantum computing.
7:01 pm
there is an initiative in the institute for molecular engineering at u chicago that has an undergraduate degree with a quantum track. we are adding to that hardware track. and there is a program -- >> that's the quantum engineering degree? >> yes. >> that you are talking about? >> yes, a quantum track of the molecular engineering degree. also a program to embody graduate students working in all areas of quantum with commercial -- with companies. and so we are participating in that. so we're trying to train other research groups so that they can start doing research in quantum. >> given the potential, it seems to me that we have to have sort of almost like a moonshot mentality about investment. you are so right about all kinds of research. if it is not steady and consistent, then, you know, we either have a brain drain.
7:02 pm
people go elsewhere. or that research grinds to a halt. but tell me a bit about some of the things you are working on, the primary and high school level. that's also under your bailiwick, too, right? >> right. at the elementary and middle school level we are looking at not doing quantum computing per se, but computer science in general because, in order to have a quantum computer scientist, you need a computer scientist first. in science anyway, if a student isn't thinking about being a scientist by sixth grade, they are statistically unlikely to become a scientist. we believe the same thing may be true for computer science. we want those initiatives early. on the physics side, what are the aspects of quantum computing that are unintuitive when you get there. one is the idea of measure. the operations look fine until you look at them. it's an issue that the
7:03 pm
measurement device perturbs the state. if you had matchbox cars and you wanted to know how fast they are going, you could put your hand out and now that stops the car. your choice of measurement actually affects the system. in quantum computing, you have no other choices. for a car, you could video it and calculate which one was faster. we don't have that opportunity in quantum computing. those sorts of things that are unintuitive can become intuitive if you give the right examples at young ages. >> great. i am out of time. i yield back. >> the chair recognizes the gentleman from illinois. >> i think, thank you for being here. i understand about 50% of the things you say. mr. brett, you say that quantum computers will allow us to solve some of the most intractable problems that exist. can you explain how doing so
7:04 pm
will benefit everyday americans? >> thank you. as we add more variabilities to them or more factors to them they become exponentially more difficult to resolve. the time that's required to solve that problem doubles every time we add a new variable to it. so we can reach a limit of their computational capacity. even with the computers and cloud computing today. for every day americans, problems like how do we optimize our financial portfolio in a 401(k)? the amount of work that's required to do that is already immense. if we want to include more factors and get the most efficiency for a portfolio. >> the scale increases exponentially. a quantum computer can help with that. we take on more complex and difficult problems and solve them in a much shorter time. >> now i am going to be honest.
7:05 pm
dr. putnam, i don't know what i'm going to say here. i am going to say it and hopefully you understand the question. okay. when you measure a qubit, it changes its value to a solid one or zero. as i understand it, which i don't, to manipulate a quantum computer, the operator needs to make measurements without a qubit observing you doing so. how do you do that and how does that match the capabilities of classic electronic computers and processors with billions of transistors? >> i feel like i should have one of the computer experts answer. this is something that occurs in physics that has been measured for many, many years. so how it's implemented becomes our greatest challenge. and there are several different ways to do it. generally, you want to be in a
7:06 pm
situation where you control the atmosphere. while it's observable in nature, it's not as controllable as dealing with information with a string of zeros and ones which just adds up in sums. i think i would like to -- you have somebody else to explain the actual technology of how it might work. dr. monroe? >> sure. first, i would like to add that you are in good company because albert einstein didn't, he never accepted quantum mechanics. he didn't think it was complete. >> so i'm like albert einstein? >> yes. [ laughter ] >> i agree. >> analogies do wonders in all of science, especially in quantum mechanics. i agree with dr. franklin's statement that finding analogies. you can teach the concepts to young children in elementary school. i totally believe that. here is an analogy for a qubit.
7:07 pm
imagine a coin. when we flip a coin, it's in a definite state all the time. if you think of it being quantum, it's heads and tails at the same time. imagine now it's in a black box and you are not looking at it. so it's both heads and tails at the same time. i want to control that coin. i want to maybe flip it. let's say it's a weighted coin so it's 90% heads and 10% tails. i want to flip the today to be 90% tails and 10% heads. we can do this from the outside world by turning the box around in a sense. we don't know what the state was. we didn't measure it. we didn't betray the quantum system but we controlled it. to dr. putnam's point, we have to keep our distance when we control it. we have to do things without looking, without looking at the -- i mean, what it means is that the system is so extremely
7:08 pm
well isolated that we don't get the information out. so a quantum computation involves manipulations like that. they can be more complicated. flip one qubit depending on the state of another, for instance, without looking, and it's possible to do that in a small set of technologies. then at the end of the day you unveil, you open the box and you measure only one state, but it could be lots and lots of bits. and that one answer can depend on exponentially many inputs in the device. as mr. brett mentioned, this can be put to use for real- world problems, logistics and so forth. >> thanks. nice work. i yield back. >> here is a large statue of albert einstein down the street in front of the state department. so you might get your statue there sometime. [ laughter ]
7:09 pm
>> okay. the chair recognizes the gentleman from kentucky for five minutes. >> thank you very much. that was a good example. i am trying to understand this and move it forward. this is in my family. i didn't get any of the genetics, but i have a nephew at the university of chicago in the physics department. he is going to cer in this summer. he is a computer science and math person working in chicago but in the financial industry. i'm trying to figure out the theory, not the theory, but the things that you are talking about that it's hard to understand to make it to the real world. mr. brett, can you tell us a little bit about what your company is doing in the financial services area? that's what my son is in, in the algorithms in hedge unfunds. our quantum computing would be an improvement over classical commuting. what difference does this make, i guess, or what is your firm doing in financial services to be better than what's currently
7:10 pm
there? >> thank you, congressman. the financial services sector is already a huge user of cloud computing technology. they are using immense amounts of come -- computational work. quantum computers will run some of the algorithms they are most efficient at. so in a mixed computer environment of financial services company will run their daily operations around compliance, portfolio optimization, understanding risk and sends some of the algorithms -- >> how is that different? >> a quantum computer allows us to solve some particular algorithms that cannot be solved on a classical machine in a timeframe.
7:11 pm
what if we need the answer today? it helps give us that speed advantage. >> it wouldn't completely replace the classical? it's too expensive? >> too expensive and/or there are some problems that quantum computers can't do. quantum computers aren't particularly good at addition or subtraction. we leave those to classical computers. quantum computers are good at optimization problems. >> okay. it's hard for my mental capacity to understand something can't do math but other things. simple math, i guess. so i am at addition and subtraction level. i am not an einstein. so dr. putnam, in your testimony, you had the problem of squaresty as one that -- scarcity that quantum computing
7:12 pm
could solve. >> thank you, congressman. often there is an enormous amount of waste in the way that we currently produce anything. this is not due to humans caring to produce waste or a problem with this in general. it's due to our inability to comprehend and to simulate and to build. the more precise we are on a molecular level, the better we are at being able to do that. the examples that i used such as fertilizer, for instance, or of material science, classical computers get very rough examples of how to actually build something and understand what is going on molecularly. the more we are able to do that in ways that quantum computing allows, the more we can explore
7:13 pm
the space of possibilities. that gives us a chance to create it. this is not possible with humans alone or with our classic computing systems. this applies to many areas that we could go on about. >> okay. >> but certainly in manufacturing. it creates an entire different way of doing manufacturing when we are precise. >> okay. i will let adam explain this to me. thanks. i understand it's a difficult concept for people not in your space to understand. but it's exciting stuff. i have 30 seconds. dr. monroe, dr. putnam mentioned about qubits, how many in quantum computers. here is a question. what is the noise ratio for qubits? how many qubits for a useful quantum computer and how many would be performing calculations? >> thanks for the question. i hope i answer to your liking.
7:14 pm
we don't know how many qubits are useful to replace conventional computers. however, a small number of about 75 or 100 qubits is enough to show certain very esoteric and narrow, maybe not useful problems can be solved that cannot be solved using conventional computers. it doesn't mean they are useful. it's sort of a proof of principle. that's going to happen soon. the question after that, once we are beyond that milepost, the idea is to find something useful. and i think the only way to find something useful is to put these devices in the hands of people that don't know or care what's inside the devices. sort of like my smartphone. i don't really want to know what's inside. and to build these devices. i use the word exotic a lot. it is exotic hardware. to build them it takes a new generation of engineers. it may be that we need hundreds, it may be we need
7:15 pm
thousands or more of these qubits for something useful. >> thank you. i yield back. >> thank you. the chair recognizes the gentleman from massachusetts. five minutes. >> thank you, mr. chairman. thank you for calling this important hearing. thank you to the panelists for being here. there is some smart people out there who are skeptical that quantum computing will become a practical reality. they say, for instance, that quantum computers are unstable for real-world problem responding. how do you respond to the skeptics and what do you see as the hurdles for a real-world application for quantum computing? >> i think if we make decisions based on that assumption we won't build a quantum computer. if we are wrong the stakes are far too high. other countries will make one. then they will be able to decrypt all of the messages -- you know, there are so many
7:16 pm
advantages if it can be realized that we don't want to be the ones who decide early and then are wrong. and we're making great strides. yes, right now quantum computers are small and error prone. so physicists like dr. monroe are working on making them more stable, larger, longer running. then there is the piece in between. it used to be that classical computers were very large in size, but very few bits and couldn't do very much. i mean, what we could do in the '80s in super computers is on your iphone now. we don't know what what can be done. we need to put the resources in to see where we can go because the stakes are too high. >> dr. monroe? >> the same technology we use to build quantum computers is lusked for quantum communication and quantum sensors these are real world
7:17 pm
applications that are deployed right now. on the sensor side, the ability to detect signals remotely via optical techniques or to detect mass, which means if you are underwater, you need to know where you are to navigate. if you are exploring for oil, you need to know what's underneath the rock. is it oil? is it water? those sensors, the limiting signal to noise in those sensors is given by quantum. quantum mechanics. and we can actually exceed those seemingly fundamental limits in some cases. i mention this because that same type of technology is used in quantum computers. so i do believe that quantum computers are the most disruptive of all these technologies, about you along the path towards that there will be other spin-ups. quantum communication is largely photonic, optic. also with single particles of
7:18 pm
light. photons. these can be used for quantum computerring in some cases. also to send data in ways that are half proof. somebody tries to observe it, they change it. they can cut the line always. they can destroy your communication, but they can't intercept it and understand it. so what does that have to do with quantum computing? if you build a big quantum computer, it will be a network. it will have probably optics that fiberize little modules. none of this hardware really exists today to couple those photons to quantum memories and qubits. i would hang my hat on quantum computing being the most disruptive, but along the way there are many other technologies related. >> dr. franklin, you got into something i wanted to ask. i have a minute and 15 seconds or so. encryption and the applications of quantum computing to encryption and the potential for it to render encryption
7:19 pm
obsolete. what's the reality of that? >> encryption is based on the idea that doing one operation is much harder than undoing it. it's a lot easier to multiply two numbers than it divide or factor a number. and so quantum -- there is a quantum computing algorithm that takes a lot of bits. that's not one of the near-term application. the numbers used to create the keys that are required to encrypt and decrypt can be broken down easily to their components. the components are necessary to decrypt. so if we get a quantum computer of that size, we are going to have to figure out new completely new encryption algorithms that use mathematical functions that a quantum computer cannot do quickly. >> and is that time horizon, is that -- can you put a time
7:20 pm
horizon on that? >> yeah. chris? >> this factoring problem is among the hardest out there. you probably need tens of thousands, maybe billions of operations. i will say the problem is so important that you need to know -- you don't want to -- you don't want a quantum computer just to break messages. you want to know when one exists. that impacts how you encrypt now. we're talking political time scales. so if a computer exists in 30 years, that could impact how you encrypt things now. you may want to be ahead of the game and change encryption standards based on when a quantum computer will exist. it's hard to predict 30 years in the future what technology will bring us. >> if you can predict what can happen tomorrow, we should hang back more. >> the chair recognizes the extra from florida for five -- the gentleman from florida for five minutes. >> thank you.
7:21 pm
thank you, mr. chairman. i'll be as brief as i can. mr. brett, in your testimony you identify three classes of applications that are possible in the near term. i know you talked about these earlier. can you briefly explain why you expect those to be the most possible in the near term? >> thank you for the question, congressman. with the earliest quantum computers like the type that chris monroe is building at the moment, the first versions of these have error correction on them. so the kind of applications we can build need to be able to accommodate errors and the potential imprecisions that come along with that. so the kind of applications that are best suited to early stage quantum computers are those which are the most tolerant or resilient to error. those are things like optimization problems, working
7:22 pm
with chemical simulation, and machine learning type problems because the kind of algorithms we run on there are based on probabilities. and so we already get a probablistic type answer from classical computers out of that, and a quantum computer best matches that. the early stage applications are those that are more probabilistic, more resistant to error. as the computers become better, we will take on the harder type problems that require error correction around that. >> thank you. this next question is for the panel. will quantum computers be something that anyone can use, which is important, or will it require highly sensitive operating environments such as the only -- only a handful would be able to operate? why don't we start from over here from afar. >> thank you, congressman. it has to be something that has
7:23 pm
user interfaces for everyone in order for it to be relevant. the physics and hardware behind it, just like the hardware and the physics behind everything else we do, will have a lot of specialists involved with it. but it's important for us, it's a challenge and important for us that this is something that is in the hands of anybody. so i think absolutely. >> so it's not going to require additional training or anything like that? >> well. >> only to the extent that everything we do requires some amount of training until it becomes so common place that it becomes natural. >> all right. very good. if you could comment on that, please. >> sure, thank you for the question. i'll be very brief. i think the answer is it will be very much like current computers. the use of current computers to
7:24 pm
program in certain languages takes some training. it will be a different type of a language, but the fact that there are individual atoms in the device at the end of the wire will be lost on the user, and it should be. they don't need to know that. they need to know the rules, the programming language, and what it can solve. i think the answer will be affirmative. >> very good. >> yeah, i think there are sort of three levels. one is hardware. we are seeing quantum cloud computations. so i think it's likely that you won't buy one and maybe have it in your pocket, but at least the cloud resources will be there. as a user you may not know that you are using a quantum algorithm. the services that you use will have programmers who have made some of the quantum -- have a combination of quantum algorithms and classical algorithms and send it to the cloud. when you do a google search, 100 programs respond. is it an airline? is it a mathematical -- you know, all these different things. in terms of the ability to
7:25 pm
program it, that's where the most work has to come in. right now the amount of expertise needed to program these is insane. i mean, it's a high level of expertise. but that's how it was when the first women programmers were given a speck of the first computer and said, here, program this. they did it from the hardware. that's where we are. it's very tied to the hardware. so we need to figure out what are those abstractions that are still useful computing-wise, but understandable to people who are the current level of a traditional computer scientist or even an application developer. >> very good. >> thank you for the question. i fully agree with my fellow panelists. we believe you shouldn't have to have a degree in quantum physics to build the computer. that's what we're doing. to do so without needing to
7:26 pm
know the intricacies at the molecular level. ibm has released an quantum computer. ibm.com/quantum. we can do a short course on quantum computing programming and build up that knowledge on what's possible and start to build those skills for the future. >> very good. i yield back, mr. chairman. >> thank you. the gentleman yields back. the chair now recognizes the gentleman from west virginia for five minutes. >> thank you, mr. chairman. again thank you for continuing to put before us in our hearings some very provocative thoughts through this disruptive series we have dealt with over the past two years. i am very curious. for me, it's one of two engineers in congress, it's exciting, the possibilities of where we might go with this.
7:27 pm
i am fascinated with it. i am also -- i'm sorry that the other side of the aisle didn't show up today. but i was curious to hear more of what kennedy was talking about, the skepticism, because when i looked into that, there is some skepticism. one of the articles i was reading a couple days ago had to do with reliability of the results. so i know from doing my own engineering, calculations, we can, at the end of the day, we know whether that result makes sense. what happens when we use a quantum computing if we get -- and i think monroe, you might have said if they are error prone. do we rely on the result? how do we question it? if we are relying on our computers to give us the answer and then we get the answer, how do we know it's wrong? or how do we know it's right?
7:28 pm
because of all the variables. you all talked about it here. >> do you want to answer that? >> yes. thank you for the question. a very good one. i think it speaks to the, so far the limited reach of what a quantum computer is useful for. there exists problems like the factoring problem. you can easily check it. 15 is equal to 5 times 3. when the 15 is a huge number, you can't do it using regular computers, but you can multiply your answer together to check and see if it worked. >> talk about encryption. >> if you factor large numbers, you can break the popular types of encryption algorithms out there now. and if you think you have a code breaker, you can check it quickly. almost all applications of quantum computers, they are either checkable against some standard or they could be better any classical approach. say, for instance, in the
7:29 pm
financial market or some logistics problem where there is a cost function. it's in real dollars and you are trying to minimize the cost subject to an uncountable number of constraints and configurations of the marketplace, for instance. if your quantum computer comes up with a potential result that has lower costs than any conventional computer could compute, then you found a different solution. >> let me, just a quick points to follow back up. i can see there is a lot more -- again, fascinating. i want to read more. this whole idea has triggered me to do a little bit more research in this as well. let's talk about the timetable. right now, yes, there are some elementary units out there. but what's the metric? what's the goal? where do we want to achieve, and how do we know whether we are there? secondly with that, what's the role of congress on this? is this just more money into
7:30 pm
research, or is this -- you talked about building plants or a facility to build these qubits. what role is government? >> well, thank you for the question again. i mentioned the idea of an national quantum initiative and the crux of that initiative is to establish indeed a small number of hub laboratories. they are not new buildings. >> hub zones or hub labs? >> quantum innovation laboratories. they could be at existing university department of energy or department of defense laboratories, collaborations with industry. hubs where students and industrial players are all in the same sand pit. there will ab small number of them. they would focus on a particular aspect of quantum information or sensing or quantum computing, develop a particular brand of qubit. it's to foster a new generation of engineers in that particular
7:31 pm
technology. industry will be able to connect more vitally with the university and a potential work force students -- >> are we trying to develop a standard qubit? >> i think it's too early to do that. i think we have several different technologies and they will find different uses. now we have a cpu in a computer, we have memory, all kinds of different components, different hardware good for different things. we will probably see that in quantum as well. >> okay. again what's the timetable? >> well, i think it depends on the application. encryption might be 30 years off. but we've got 50 qubit machines now. they are growing. so near-term applications like optimization may be five years. the hardware is coming along quickly, i think, and some software. this is the first i have heard of a software company. i'm really excited.
7:32 pm
there is software that needs to be created so can be modified to use thener- term hardware so we can close that gap between the assumptions of the software and the realities of the hardware. >> thank you. i yield. >> thank you. the gentleman yields back. the chair recognizes the gentleman from indiana. five minutes. >> thank you for being here. fascinating subject. i was a surgeon before, so i'm kind of a scientist. i am interested in this. my daughter is a sophomore at cornell as a computer science, so she is obviously. i will take a little different pathway here on questioning and stay away from the technical stuff and just go towards research funding. and i was on a committee before that had jurisdiction over national science foundation, so i'm from indiana. i went to all the universities and talked to the nsf funded researchers. the thing that i found is, first of all, i support that. i am a big supporter of research.
7:33 pm
one thing i found is if i said, hey, tell me why what you are doing should continue to get funding from the national science foundation, just a simple question, right? i found probably 90% of the people that i spoke to couldn't, in a really tight way, explain that. and for me, you know, they can ex main it in a complex way. i am like, yeah, i get it. people like me have to explain this to 700,000 people that we represent in a way that, if we are going to justify federal dollars and taxpayer dollars, we have to give a so-called elevator speech and say one example, this is four or five years ago that was kind of in the press, was about a funded researcher. this is not a criticism. they were having seniors play video games. and so it got in the press and people said, why would you fund
7:34 pm
that? and as it turns out, it was alzheimer's research. see what i'm saying. very valid, very important research. to try to explain that, you know, when it's written in the line, you know, government funds video game, you know, making -- having people be better video game player doesn't play very well. and so people like me have a hard time explaining that. so i guess what i'm getting at is, and i guess this will be primarily for the people from the universities, is what's your pitch for more funding for quantum computing? >> i know that's a -- you have already explained it to me, and i get it. but if we are going to explain it to the broader members of congress and our constituents, how do we explain that, why we should do that? does that make sense? >> yes, it does. thank you for the question, congressman. i did speak at length about these very targeted type hubs.
7:35 pm
it should be self-evident what these are about. they are more technology centers. there must be an undercurrent of foundational research. this is something the national science foundation, they are a very special agency in that regard. fundamental research is very inefficient. we can never tell what's around the corner. and you can never predict what is going to hit -- >> yeah, you don't know what you don't know, right? >> that's right. >> the science foundation takes all commerce, and they -- all comeers. there may be quantum technologies that don't exist now. maybe in ten years we see that. too bad it's inefficient. but the home runs are far reaching. this field will probably rely on those in the coming decades.
7:36 pm
>> dr. franklin. >> yeah. i mean, i think it depends how long you are in the elevator. i think the pitch for quantum computers starts with the killer apps of drug design for alzheimer's, right? it's projected 40% of the medicaid budget is going towards alzheimer's by 2040. i mean, those are real problems. if we could model the molecules and figure out how nitrogen gets fixed and put into fertilizer, we could have lower energy food production. these are big deals, right? and those are things that can't be done with classical computerring. then you have to tie the researchers to those problems. we are at the cusp of commercialization. it might be an appropriate time for even the n sf funding to look at the broader impact more. our group is making tools that everyone can use. so that's something that we can
7:37 pm
hang on to. >> okay- >> the other thing i'm interested in is technology transfer, obviously, because that is, as you know, a huge problem not only in this area, but across the research field. i mean, what percentage of research goes, you know, that is potentially commercially useful, it goes into a black hole. i know i'm short on time. maybe mr. brett, you commented -- i mean, how we can do better on technology transfer. it's a pretty big problem really. >> thank you, congressman. we agree. as a small business that is looking to commercialize some of these innovations, how do we get access to some of the great work being done at of the universities . >> it's proprietary sometimes, right? they put their research out there, they are worried somebody will steal it so to speak, right? >> we found an approach that has been particularly successful for us is being able to partner with the
7:38 pm
universities on research grants. our business also to participate in the collaboration of that research and contribute to the science and publication around that and share some of that intellectual property on a joint project today. that cross between the commercial sector and research sector working together on funded proposals will enable a lot of that technology transfer. >> my time is up. i yield back. >> the gentleman yields back. first of all, i want to thank our panel for being here today. one of the great things about serving on this committee, it's like looking over the horizon five, ten years. we hear it here first. we want to make sure our nation is on the cutting edge. i have a former air force pilot, west point grad, a cardiothoracic surgeon here. they are not limited.
7:39 pm
but what you gave us today was very -- very, very informative. we have to make sure, as we go forward, as a committee, that we are making the right decisions as we go on. yes? >> yeah. so i just want to thank you all. i'll finish up the ending. but i'll let the lady go now. >> china is building a $10 billion quantum lab right now, and they expect to be finished by 2020. and the eu is investing about $2 billion in advanced quantum technology. i think one of the answers, in terms of why we should be serious about making investments, maybe decryption is an -- and encryption is some decades away. but from a national security perspective, i think that there
7:40 pm
are a lot of reasons that we should take this seriously, make the investments. of course, all the practical things about agriculture and pharmaceuticals, et cetera, is very, very important. disease cures. but it seems to me that despite maybe some skepticism, there is enough evidence right now that this really ought to be an important priority. i just want to thank you very much. you really did enlighten me. thank you. >> thank you. and we have no further members that are going to be asking questions today. pursuant to committee rules, i remind members they have ten business days to submit additional question forts record. -- for the record. and without objection, the subcommittee will stand adjourned. thank you very much for attending today.
7:41 pm
7:42 pm
7:43 pm
washington journal live every day with news and policy issues that impact you. coming up saturday morning jillian berman will join us to discuss the cost and value of college education as student debt rises. eric kelderman will talk about changes to policy under betsy devos. watch the washington journal live at 7 eastern saturday morning. join the discussion. some of the program highlights for this memorial day weekend. saturday on c-span at 9:30 eastern the monk debate. is political correctness a threat to free speech or a positive force for social justice? on book tv at 11 p.m. eastern jon meacham, author of the soul of america, the battle for our better angels.
7:44 pm
and american history tv on c- span 3 at 10 p.m. eastern archival films of world war i. sunday on c-span at 6:30 p.m. former new jersey governor chris christie at the university of chicago institute on politics. on book 2 on c-span 2 on afterwards, former national intelligence director james clapper. sunday on american history tv on c-span 3 american artifacts, a tour of the battlefield and american cemetery in france. memorial day on c-span at 11:00 a.m. eastern live coverage of the wreath-laying ceremony at the tomb of the unknown soldier. on book tv on c-span 2 at 8:30 p.m. eastern in tenth with with david

84 Views

info Stream Only

Uploaded by TV Archive on