.svg)
Podcast with Tom Wong, quantum information scientist and author
My guest today is Tom Wong, physicist, quantum information scientist and author of the new book: “Introduction to classical and quantum computing”. Tom and I talk about teaching quantum computing to high school and undergraduate college students, about his cutting-edge research in quantum algorithms, and much more.
Listen to additional podcasts here
THE FULL TRANSCRIPT IS BELOW
Yuval: Hello Tom, and thanks for joining me today.
Tom: Hello. Thank you for having me.
Yuval: So, who are you and what do you do?
Tom: So I am a physics professor at Creighton University in Omaha, Nebraska, and I primarily work with undergrads doing quantum computing research, and teaching quantum computing.
Yuval: And you have a new book that I think as I saw today is number one on the Amazon list for quantum computing books?
Tom: It's number one for new releases, obviously it's not going to be number one compared to all of the seminal titans in our field. But yeah, I do have a new textbook that's based on a class that I've been teaching at Creighton to undergrads that's an introductory quantum computing course. An interesting thing about this is that the only prerequisite is trigonometry. So a lot of the educational materials that exist with quantum computing and textbooks, they're more targeted for a graduate-level student, but this is made really for freshman, sophomore-level students in undergrad, and maybe even high school students that are more advanced.
Yuval: I downloaded a copy and I read most of it, and I think it's lovely. I do hope I know most of it, although I haven't taken any of your exams quite yet. How early do you think this can be taught? Would you teach it to 9th graders? Would you teach it to 11th graders? I mean, where would you recommend people start to get into quantum?
Tom: So, I guess there's kind of two questions with that. One is at what level could I teach the material in my textbook? And the other question is at what level could you teach quantum information science more broadly? So with my textbook, I mentioned that the prerequisite is trigonometry. So basically once students are familiar with the unit circle, they could tackle all the material in my textbook, because my book does cover the more advanced math that they'll need. So it covers linear algebra and reviews complex numbers, and things like that. Maybe I should mention for the listeners that actually made this textbook free, that there's a free PDF of it on my website at thomaswong.net. And then for people who want a more affordable print copy, I just uploaded it to Amazon. So you can buy a cheap print copy as well. But again, for the PDF, it's just free on my website, so anyone can download it.
With that there are plenty of high schoolers who take trigonometry, who could then learn quantum computing. Beyond my textbook, I think the basic ideas of quantum information actually can be taught without all the mathematics, even to kids. And I think there are actually pushes to do this. So for example, there's this initiative in the US called the Q-12 Educational Partnership. I think the website is q12education.org . And it's basically people recognizing that there's a need to have a future quantum workforce. And to do that, you need to introduce ideas in quantum information science at younger ages. That way students are aware that the field exists, and then they can think, Oh, when I grow up, I want to be a quantum information scientist, because a lot of times students go into a particular field simply because that's what they've heard about from their parents or from their teachers or school counselors. And if quantum computing is not on people's radars, they won't know that these jobs even exist.
Yuval: You've been teaching this for a number of years now. So what's the most difficult of concept for your students to grasp?
Tom: Ooh, that's a really tough question, I think. I don't want to say that there's no tough concept to grasp, because then it sounds like I'm bragging or something. But I think some of it depends on the type of course that you teach, because my course is not a conceptual course. I actually teach them the math that they don't know. Because we actually do the math, I think students understand these very quantum concepts like quantum entanglement and super position and things like that, because they've seen how it actually works with the math. And I think what makes some of these concepts hard to understand to a lay audience is that you're trying to use analogies to describe them instead of just knowing what they actually are. And I think that's when things like entanglement can become very confusing. And so I think, at least with the topics that I cover in my classroom, which again are introductory topics, I'm not teaching a graduate level course. I think that the basics are actually accessible to students, and even the basics of entanglement, and superposition and all these quantum concepts, they're actually not too bad.
And I think that might be a very good takeaway actually, which is a lot of times quantum has a reputation of being mysterious or that only the brightest minds in the world can understand it, and can comprehend it, and can contribute to this field. And I want to break that stereotype, I really think that anyone can understand and contribute to quantum. And so in some sense, we've had marketing issues at the field because we've marketed as something that is spooky. But in fact, just like anything you learn, if you take all the necessary steps in order to understand it, it shouldn't be spooky anymore, because the job of scientists is for the things we're studying to no longer be spooky because you understand it. I mean, if you have a little kid and you talk to them about even how to multiply numbers, when all they know is how to count on their fingers, multiplication sounds something that can be intimidating and that they could never understand.
But of course, once they go through school and learn addition and subtraction, and once they get to multiplication, it's like, Oh it's not too bad, because they've taken the steps necessary to grasp it. And so I think we need to convey that about quantum computing. That maybe for where currently are it is a little bit of a reach, but you just need a couple of steps filled in and then it's something that can be grasped.
Yuval: Let's make a sharp turn and actually go to sort of modern day, and some of the more advanced things that you're working on outside the book, and then I do want to come back to the book. People are familiar with the famous quantum computing algorithms, Shor's and Grover's, and Deutsch–Jozsa and so on. Why are there so few, I mean, wouldn't you expect that by now there'd be 30 or 50 major quantum algorithms?
Tom: I think a big reason is actually that we don't have big enough quantum computers for people to play around with and to try. And what I mean by that is if you look at the evolution of classical computing. Back when classical computers were first being invented, there were certain problems that people had in mind for what they would do with these classical computers. And because of that, I think there's some famous statement by a CEO at the time who said that the world market for computers is five, because they had such a limited set of problems in mind for what classical computers could do. But once computers became available to everyone and more people could be involved, people started coming up with all sorts of uses, such as this podcast that were recording over the internet or even a lot of classical machine learning, a lot of it.
The reason why we use machine learning is because it just works. People have tried it and it works and there's not necessarily the most rigorous theoretical underpinning for why machine learning is so successful. I mean, there's some like general senses for why, but in terms of the rigorous mathematical computer science type proof, that doesn't even exist with classical machine learning. And so I think in the same way with quantum computing, yeah, we have those handful of kind of big problems that we could solve with quantum computers, but really I think most of the applications won't be discovered until the quantum computers are more available for people to play with, and you just try it, try something and see if it works. And then later on the mathematicians and theoretical computer scientists can fill in the theoretical understanding for why it works.
Yuval: What are you working on? What kind of algorithms are you investigating?
Tom: I work primarily with quantum algorithms that are based on the quantum versions of random walks. So random walk is just when something hops around randomly. And it turns out that this process is very useful for classical algorithms, and in the same way it turns out to be a very useful way to design quantum algorithms. And in particular, I look at quantum search algorithms. So you imagine you have some type of network, and you have your quantum particle that's jumping around on this network in a super position because it's quantum, and it's trying to look for a particular node. And it turns out that depending on the structure of that network, a quantum computer might be able to search quickly or slowly. And so I investigate a lot about what different properties of the network might be important for a quantum computer to search quickly.
Yuval: Is that related it in any way to Monte Carlo algorithm that also have some sort of a random walk built into them?
Tom: It is related. Yeah. You can think of a quantum walk as being like a quantum version of a Monte Carlo algorithm or like a Markov chain.
Yuval: Understood. So when I look at the book and you explain these algorithms. Well, here's how I build an Oracle for a few qubits and here's how I do this, and so on. What do you think happens when people, let's say I read the book, I did all the exams, I know what's there. What's the gap between the book and actually creating something truly useful in a corporate quantum computing environment. What's next thing.. or what do people have to go through to get there?
Tom: Yeah. Well, one of the nice things about where quantum computing currently is, is that there's a lot of educational resources. So I don't claim that my book is the only one out there. There are a lot of resources available that even a lot of quantum computing companies have put out, and they have whole lessons and even free eTextbooks on how to use their devices for potential optimization problems and things like that. And so there is a lot of work right now, and a lot of material on how one might be able to use quantum computers for industrial problems. Currently the quantum computers aren't big enough as to solve any of these problems usefully beyond or better than what our existing traditional computers can do. But the idea is to start experimenting with these small quantum computers, that way when we have bigger ones we can hopefully apply them to actually solve problems that we can't currently solve. And so I would definitely point people to those types of resources.
Yuval: And in terms of your research, do you see it having commercial applicability?
Tom: There could be, because searching is a very important problem. That's why Grover's algorithm is so celebrated because, even though it's a quadratic speedup not exponential speedup, it's a very universal type of problem. And so you can almost think of my research as looking a little bit more into the details of Grover's algorithm. Like what happens if your data is structured in a way where you can't just jump from one piece of data, one node to another node directly where it's arranged in some type of structure. See, I think searching is a very universal problem. In terms of when it will be commercially viable? It might be awhile because quantum computers it's a very tough engineering problem. And also our classical computers are actually just very, very good. And so, it's going to take a little bit of time I think for quantum computers to start to do things that classical computers can't with just a more modest polynomial speedup.
Yuval: So, apologies to the random walk in my questions. Let's go back to the book and some of the basic concepts. I think that when people learn about entanglement, they get sort of the value in terms of quantum communications or key distribution, and oh, if I change one, then the other changes so you would know and so on and so on. How would you explain to someone the value of entanglement in the computing relevance of quantum?
Tom: Yeah. So in our book, we talk about what entanglement is. So entangled state is state where all your qubits are entangled or mixed together in some way, right? Where if you measure one qubit, it affects the other qubits, which is the opposite of a state where that is not true, which is called a product state. So in a product state, your qubits can be thought as being individual qubits, where you can interact with one qubit without affecting the other qubits. So if there's no entanglement, meaning you have a product state, you can actually simulate a product state efficiently using classical computers. And it's because, I'm guessing this is a more advanced audience for your podcast, so people might know a little bit about quantum computing. Is because the number of amplitudes that you have to keep track of for product state actually grows linearly with the number of qubits as opposed to exponentially.
And so because of that, a classical computer can store the entire quantum state for product state efficiently, and it can act on it efficiently as well. So basically if there's no entanglement, a quantum computer is only as good as a classical computer in that regard, because a classical computer can do everything that that quantum computer can do. So if you want a speedup, you need to use entanglement essentially. So I think that is the big motivation for why it's important in computation, you basically need it if you want to do anything better than a classical computer.
Yuval: As you look at quantum applications today, and going back to the commercial use, obviously there's a lot of hype and fear around Shor with regards to, oh, I'm going to break the world's financial system, but that's still a few years away. When you talk to industry colleagues, what is the algorithm or classical algorithms that they use most more than other in their work?
Tom: I think it's going to depend a lot on the industry of course. I mean, if you're talking about problems that are more around quantum simulation, quantum chemistry, things like that, then basically trying to find ground states so that's a very general thing. And then a lot of other industries it's basically solving differential equations, and in finance that's true, for different financial models, things like that. In other industries it's how to optimize things, how to optimize your flight schedules, things like that. So I think there's a lot of variation, actually, depending on the industry, which also makes it very interesting. I mean, it's like asking with a traditional computer, what are classical computers used most often for? It's like everything. So yeah. I don't know if I can give you one specific area.
Yuval: That's fantastic. So Tom, how can people get in touch with you to learn more about you, and your work in the book?
Tom: So my website, thomaswong.net is the best place where you can see all my research papers. You can get a link to download my free textbook. And at the bottom is also my contact information on my website.
Yuval: That's perfect. Thanks so much for joining me today.
Tom: You're welcome. Thank you for having me.
My guest today is Tom Wong, physicist, quantum information scientist and author of the new book: “Introduction to classical and quantum computing”. Tom and I talk about teaching quantum computing to high school and undergraduate college students, about his cutting-edge research in quantum algorithms, and much more.
Listen to additional podcasts here
THE FULL TRANSCRIPT IS BELOW
Yuval: Hello Tom, and thanks for joining me today.
Tom: Hello. Thank you for having me.
Yuval: So, who are you and what do you do?
Tom: So I am a physics professor at Creighton University in Omaha, Nebraska, and I primarily work with undergrads doing quantum computing research, and teaching quantum computing.
Yuval: And you have a new book that I think as I saw today is number one on the Amazon list for quantum computing books?
Tom: It's number one for new releases, obviously it's not going to be number one compared to all of the seminal titans in our field. But yeah, I do have a new textbook that's based on a class that I've been teaching at Creighton to undergrads that's an introductory quantum computing course. An interesting thing about this is that the only prerequisite is trigonometry. So a lot of the educational materials that exist with quantum computing and textbooks, they're more targeted for a graduate-level student, but this is made really for freshman, sophomore-level students in undergrad, and maybe even high school students that are more advanced.
Yuval: I downloaded a copy and I read most of it, and I think it's lovely. I do hope I know most of it, although I haven't taken any of your exams quite yet. How early do you think this can be taught? Would you teach it to 9th graders? Would you teach it to 11th graders? I mean, where would you recommend people start to get into quantum?
Tom: So, I guess there's kind of two questions with that. One is at what level could I teach the material in my textbook? And the other question is at what level could you teach quantum information science more broadly? So with my textbook, I mentioned that the prerequisite is trigonometry. So basically once students are familiar with the unit circle, they could tackle all the material in my textbook, because my book does cover the more advanced math that they'll need. So it covers linear algebra and reviews complex numbers, and things like that. Maybe I should mention for the listeners that actually made this textbook free, that there's a free PDF of it on my website at thomaswong.net. And then for people who want a more affordable print copy, I just uploaded it to Amazon. So you can buy a cheap print copy as well. But again, for the PDF, it's just free on my website, so anyone can download it.
With that there are plenty of high schoolers who take trigonometry, who could then learn quantum computing. Beyond my textbook, I think the basic ideas of quantum information actually can be taught without all the mathematics, even to kids. And I think there are actually pushes to do this. So for example, there's this initiative in the US called the Q-12 Educational Partnership. I think the website is q12education.org . And it's basically people recognizing that there's a need to have a future quantum workforce. And to do that, you need to introduce ideas in quantum information science at younger ages. That way students are aware that the field exists, and then they can think, Oh, when I grow up, I want to be a quantum information scientist, because a lot of times students go into a particular field simply because that's what they've heard about from their parents or from their teachers or school counselors. And if quantum computing is not on people's radars, they won't know that these jobs even exist.
Yuval: You've been teaching this for a number of years now. So what's the most difficult of concept for your students to grasp?
Tom: Ooh, that's a really tough question, I think. I don't want to say that there's no tough concept to grasp, because then it sounds like I'm bragging or something. But I think some of it depends on the type of course that you teach, because my course is not a conceptual course. I actually teach them the math that they don't know. Because we actually do the math, I think students understand these very quantum concepts like quantum entanglement and super position and things like that, because they've seen how it actually works with the math. And I think what makes some of these concepts hard to understand to a lay audience is that you're trying to use analogies to describe them instead of just knowing what they actually are. And I think that's when things like entanglement can become very confusing. And so I think, at least with the topics that I cover in my classroom, which again are introductory topics, I'm not teaching a graduate level course. I think that the basics are actually accessible to students, and even the basics of entanglement, and superposition and all these quantum concepts, they're actually not too bad.
And I think that might be a very good takeaway actually, which is a lot of times quantum has a reputation of being mysterious or that only the brightest minds in the world can understand it, and can comprehend it, and can contribute to this field. And I want to break that stereotype, I really think that anyone can understand and contribute to quantum. And so in some sense, we've had marketing issues at the field because we've marketed as something that is spooky. But in fact, just like anything you learn, if you take all the necessary steps in order to understand it, it shouldn't be spooky anymore, because the job of scientists is for the things we're studying to no longer be spooky because you understand it. I mean, if you have a little kid and you talk to them about even how to multiply numbers, when all they know is how to count on their fingers, multiplication sounds something that can be intimidating and that they could never understand.
But of course, once they go through school and learn addition and subtraction, and once they get to multiplication, it's like, Oh it's not too bad, because they've taken the steps necessary to grasp it. And so I think we need to convey that about quantum computing. That maybe for where currently are it is a little bit of a reach, but you just need a couple of steps filled in and then it's something that can be grasped.
Yuval: Let's make a sharp turn and actually go to sort of modern day, and some of the more advanced things that you're working on outside the book, and then I do want to come back to the book. People are familiar with the famous quantum computing algorithms, Shor's and Grover's, and Deutsch–Jozsa and so on. Why are there so few, I mean, wouldn't you expect that by now there'd be 30 or 50 major quantum algorithms?
Tom: I think a big reason is actually that we don't have big enough quantum computers for people to play around with and to try. And what I mean by that is if you look at the evolution of classical computing. Back when classical computers were first being invented, there were certain problems that people had in mind for what they would do with these classical computers. And because of that, I think there's some famous statement by a CEO at the time who said that the world market for computers is five, because they had such a limited set of problems in mind for what classical computers could do. But once computers became available to everyone and more people could be involved, people started coming up with all sorts of uses, such as this podcast that were recording over the internet or even a lot of classical machine learning, a lot of it.
The reason why we use machine learning is because it just works. People have tried it and it works and there's not necessarily the most rigorous theoretical underpinning for why machine learning is so successful. I mean, there's some like general senses for why, but in terms of the rigorous mathematical computer science type proof, that doesn't even exist with classical machine learning. And so I think in the same way with quantum computing, yeah, we have those handful of kind of big problems that we could solve with quantum computers, but really I think most of the applications won't be discovered until the quantum computers are more available for people to play with, and you just try it, try something and see if it works. And then later on the mathematicians and theoretical computer scientists can fill in the theoretical understanding for why it works.
Yuval: What are you working on? What kind of algorithms are you investigating?
Tom: I work primarily with quantum algorithms that are based on the quantum versions of random walks. So random walk is just when something hops around randomly. And it turns out that this process is very useful for classical algorithms, and in the same way it turns out to be a very useful way to design quantum algorithms. And in particular, I look at quantum search algorithms. So you imagine you have some type of network, and you have your quantum particle that's jumping around on this network in a super position because it's quantum, and it's trying to look for a particular node. And it turns out that depending on the structure of that network, a quantum computer might be able to search quickly or slowly. And so I investigate a lot about what different properties of the network might be important for a quantum computer to search quickly.
Yuval: Is that related it in any way to Monte Carlo algorithm that also have some sort of a random walk built into them?
Tom: It is related. Yeah. You can think of a quantum walk as being like a quantum version of a Monte Carlo algorithm or like a Markov chain.
Yuval: Understood. So when I look at the book and you explain these algorithms. Well, here's how I build an Oracle for a few qubits and here's how I do this, and so on. What do you think happens when people, let's say I read the book, I did all the exams, I know what's there. What's the gap between the book and actually creating something truly useful in a corporate quantum computing environment. What's next thing.. or what do people have to go through to get there?
Tom: Yeah. Well, one of the nice things about where quantum computing currently is, is that there's a lot of educational resources. So I don't claim that my book is the only one out there. There are a lot of resources available that even a lot of quantum computing companies have put out, and they have whole lessons and even free eTextbooks on how to use their devices for potential optimization problems and things like that. And so there is a lot of work right now, and a lot of material on how one might be able to use quantum computers for industrial problems. Currently the quantum computers aren't big enough as to solve any of these problems usefully beyond or better than what our existing traditional computers can do. But the idea is to start experimenting with these small quantum computers, that way when we have bigger ones we can hopefully apply them to actually solve problems that we can't currently solve. And so I would definitely point people to those types of resources.
Yuval: And in terms of your research, do you see it having commercial applicability?
Tom: There could be, because searching is a very important problem. That's why Grover's algorithm is so celebrated because, even though it's a quadratic speedup not exponential speedup, it's a very universal type of problem. And so you can almost think of my research as looking a little bit more into the details of Grover's algorithm. Like what happens if your data is structured in a way where you can't just jump from one piece of data, one node to another node directly where it's arranged in some type of structure. See, I think searching is a very universal problem. In terms of when it will be commercially viable? It might be awhile because quantum computers it's a very tough engineering problem. And also our classical computers are actually just very, very good. And so, it's going to take a little bit of time I think for quantum computers to start to do things that classical computers can't with just a more modest polynomial speedup.
Yuval: So, apologies to the random walk in my questions. Let's go back to the book and some of the basic concepts. I think that when people learn about entanglement, they get sort of the value in terms of quantum communications or key distribution, and oh, if I change one, then the other changes so you would know and so on and so on. How would you explain to someone the value of entanglement in the computing relevance of quantum?
Tom: Yeah. So in our book, we talk about what entanglement is. So entangled state is state where all your qubits are entangled or mixed together in some way, right? Where if you measure one qubit, it affects the other qubits, which is the opposite of a state where that is not true, which is called a product state. So in a product state, your qubits can be thought as being individual qubits, where you can interact with one qubit without affecting the other qubits. So if there's no entanglement, meaning you have a product state, you can actually simulate a product state efficiently using classical computers. And it's because, I'm guessing this is a more advanced audience for your podcast, so people might know a little bit about quantum computing. Is because the number of amplitudes that you have to keep track of for product state actually grows linearly with the number of qubits as opposed to exponentially.
And so because of that, a classical computer can store the entire quantum state for product state efficiently, and it can act on it efficiently as well. So basically if there's no entanglement, a quantum computer is only as good as a classical computer in that regard, because a classical computer can do everything that that quantum computer can do. So if you want a speedup, you need to use entanglement essentially. So I think that is the big motivation for why it's important in computation, you basically need it if you want to do anything better than a classical computer.
Yuval: As you look at quantum applications today, and going back to the commercial use, obviously there's a lot of hype and fear around Shor with regards to, oh, I'm going to break the world's financial system, but that's still a few years away. When you talk to industry colleagues, what is the algorithm or classical algorithms that they use most more than other in their work?
Tom: I think it's going to depend a lot on the industry of course. I mean, if you're talking about problems that are more around quantum simulation, quantum chemistry, things like that, then basically trying to find ground states so that's a very general thing. And then a lot of other industries it's basically solving differential equations, and in finance that's true, for different financial models, things like that. In other industries it's how to optimize things, how to optimize your flight schedules, things like that. So I think there's a lot of variation, actually, depending on the industry, which also makes it very interesting. I mean, it's like asking with a traditional computer, what are classical computers used most often for? It's like everything. So yeah. I don't know if I can give you one specific area.
Yuval: That's fantastic. So Tom, how can people get in touch with you to learn more about you, and your work in the book?
Tom: So my website, thomaswong.net is the best place where you can see all my research papers. You can get a link to download my free textbook. And at the bottom is also my contact information on my website.
Yuval: That's perfect. Thanks so much for joining me today.
Tom: You're welcome. Thank you for having me.
About "The Qubit Guy's Podcast"
Hosted by The Qubit Guy (Yuval Boger, our Chief Marketing Officer), the podcast hosts thought leaders in quantum computing to discuss business and technical questions that impact the quantum computing ecosystem. Our guests provide interesting insights about quantum computer software and algorithm, quantum computer hardware, key applications for quantum computing, market studies of the quantum industry and more.
If you would like to suggest a guest for the podcast, please contact us.
