August 12 Physics AMA with Daniel Sank: question pool

Question

I am Daniel Sank, guest for the Physics Stack Exchange AMA taking place in the hbar chat room on August 12 at 16:00 UTC [a]. I am looking forward to discussing my experience in the USA physics education system from lower education schools up through the PhD program, my research in experimental quantum computing, the difficulties and importance of physics pedagogy, and anything else you'd like to talk about.

Post questions you'd like answered below

The AMA will be informal, but to get an idea of what others would like to talk about, please post questions you'd like answered as responses to this meta post. See, for example, the questions posted for Slereah's AMA.

Education and work history

I was born and raised in the United States. I went to mostly public schools, with a three year detour into private school for grades 3, 4, and 5. I did my bachelor's degree in physics at Yale University and my PhD at University of California Santa Barbara in the lab of John Martinis where I worked on precision measurement of magnetic noise and on quantum state measurement, both in superconducting qubits.

I now continue work in superconducting qubits at the Google Quantum Artificial Intelligence Lab. In this position I continue to work on state measurement, as well as several other aspects of building a quantum computer.

Technical interests

• Noise

  • As an experimentalist, I have found it important to understand noise in signal processing contexts
  • For obvious reasons, I'm also interested in the noise which limits superconducting qubit performance.
  • I have found that a true understanding of noise is absolutely essential to my work, and was mostly omitted from my education! It all started when I took the statistical mechanics course as an undergraduate, and I like to answer noise related statistical mechanics questions on the main site.
  • I particularly enjoy random process problems. See this Stack Exchange post for an awesome example (note the amazing variety of answers!).

• Programming

  • Messaging protocols, and particularly capability based protocols. I am currently working on cappy.

  • Dissemination of good coding practices and shared code within the scientific community. I am a contributing member of the LabRAD project, an RPC system in use in use by at least fourteen labs throughout the world.

  • I'm happy to talk about any of my projects on github or anything else you'd like to discuss.

• Pedagogy

  • I enjoy understanding supposedly complicated topics in simple yet rigorously correct ways. Here's and example from the Physics.SE main site.
  • I once gave a two-part lecture on the Fourier transform for my undergraduate journal club. The positive responses from that form one of the best memories of my entire life and is one of the events that convinced me that time spent preparing very good pedagogical material is worthwhile. Those lectures are written up here, but in need of much work (collaborators welcome).
  • Mostly I want to help improve our standards for scientific pedagogy and would enjoy questions/discussion on that topic.

Nontechnical interests

• Music. I play the classical guitar.

• Plants and animals. At present I have many succulents, a few orchids, and a beautiful tri-color spiderwort. I am very interested in octopuses, because they are awesome.

• Cooking. I have invented the best vegetarian chili known to man-kind. I also enjoy talking about pizza crust and the brewing of honey wine.

• Backpacking. So far I've done two trips in Yosemite, one in the Cascades, one in Sequoia, one in the Sierra mountains, and one in the Catskills.

• Super Smash Bros. Melee. Here is one example of why. If you don't have the time to watch the entire game linked there, just watch this part.

[a] For USA users that's 9:00 Pacific, 10:00 Mountain, 11:00 Central, and 12:00 Eastern.

Answer 1

you say

I have found that a true understanding of noise is absolutely essential to my work, and was mostly omitted from my education!

I am facing a similar problem! I would like to ask: How did you acquire your knowledge about this branch of statistical mechanics? Would you be able to provide any good references? Most standard statistical mechanics books seem to be omitting this topic largely. I found some stuff in Altland and Simons, but would appreciate your input and potentially a systematic structuring of how to learn about the topic.

Thanks!

Answer 2

What inspired or motivated you to become a physicist? Were any teachers or family particularly important? or did you discover science yourself through books, TV and the internet? Were there ever any barriers - financial or social, or other - to your career or education? If so, how did you overcome them?

Answer 3

If I could attend to your interview, I have some ideas on subjects that would interested me and hopefully others :

1)Education Some of your ideas for physics in schools

Your opinion on the current existing system of higher education over the world. What do you believe for the way the Masters and PhDs are working in physics. For example, do you think that they are too much directed towards the needs of the working sector or not. Any ideas and experiences on the field. Also, maybe a brief discussion for the PhD program of the USA, do you think it works, should it be like this, is it a good opportunity for someone to study physics in the US and why?

Most importantly, as you mentioned for the noise in systems an that it was completely a missing subject in your education, are there any thoughts on reshaping- reformulating the universities programs and towards what directions- what classes should be omitted and what should be added?

In the end to put it compact, do you have any ideas on the didactics of physics?

2)Physics This is also important, unfortunately I do not have the knowledge to approach your scientific interests. But, the reason I wrote this post is mainly because I think I can follow a discussion by you of a new paper I found on chaotic behaviour and entanglement- if I am not mistaken you participated on this work. I t would be very interesting to learn something about this since the possibilities this work opens for future understanding of nature are, in my opinion, exceptional. Here is a link: http://phys.org/news/2016-07-blur-line-classical-quantum-physics.html?utm_source=nwletter&utm_medium=email&utm_campaign=weekly-nwletter

I hope I can attend to this meeting( I missed yuggibs).

Thank you.

Answer 4

Another question about noise: what is your view on the current state of a universal theory of noise vs detailed process dependent theories of noise, particularly concerning of 1/f noise?

Answer 5

Automating a big part of tuition through interactive e-books. Is it possible?

Current tuition system involves a professor explaining, motivating, guiding students, and students reading at home.

There are some big issues with that:

• Pace:
  When there are 20-50 students in a classroom being taught the same subject simultaneously, it is impossible to match everyone's pace of learning. Some people learn much faster, and being taught at a slower pace is a waste of their time.
• Requires physical presence in a physical room:
  While it might seem negligible, the cost of moving everyday into and out of the room, the cost of building the room, the cost of paying a professor every day for 40 years to go into the room and lecture.
• Writing on a backboard:
  It has its benefits, but it is also time-consuming. It might seem small, but multiplying the repetitive actions every day by 40 years or so becomes huge. One wonders why most of this process hasn't been automated by everyone.
• Paper books:
  Not everything can be understood at school (or university), so students need to read a book at home. A book that often has errors, or concepts that can be explained in a much better way.

And some irreplaceable benefits:

• Motivation:
  Along with being motivated by a professor that loves his subject, there is also pressure for being on time when it comes to homework; both due to knowing your professor expects the homework and also due to (inevitably) comparing yourself to other students.
• Guidance:
  Not everything can be written in a book, or can be easily found. A student can ask a professor for more reliable advice than looking up specific subjects on their own.

(The above lists are not exhaustive of course)

---

The problem is that for centuries there has been a waste of human resources during tuition, mostly due to lack of tools.

Those tools now exist:

• Jupyter (formerly known as ipython)
• StackExchange.... no really. The model of "contribute improvements and we'll reward you with imaginary trophies" has proven to be extremely successful
• There is also Mathematica which allows easily creating interactive e-books (but then again the cost of proprietary software is a big no for some institutions)

Question:

Would it be possible to have (university) students gradually rewrite all books in electronic, interactive form, guided by their professors? Their reward: experience on technologies like LateX, Python, and a small bonus to their grades (that's the imaginary trophy), and most importantly, a better book for their peers.

No more errors (or at least they'll get fixed extremely fast), concepts can be 3D animated and interactively explained, at the appropriate pace for each student, no more thousands of man-hours spent by professors repeating the same thing over and over again.

Searching through a book will be as easy as pressing ctr+f. Going to another book could be as easy as clicking a hyperlink.

Professor-student interaction will not be replaced, it will simply be made more efficient. Instead of repeating something an interactive e-book can teach, professors and students can spend more "quality time" on things a book can't provide. There will also be more time for experiments.

Answer 6

Recently I and others are playing the science simulation game Quantum moves. As we solve the levels and read about the nature article we have some questions about quantum computing in the technical context of manipulating atoms with optical tweezers (which is expanded by question 1 into a more general question that ask about the experimental aspects of quantum computing and the various type of problem in the implementation of a quantum computer).

1. Besides the Atomic Transport problem mentioned in the nature article that talked about the data obtained from this game, what other technical aspects in terms of manipulation via optical tweezers is important in the context of realising a quantum computer?

2. (A more specific version of question 1) On the final level of the game "Get it Together" the aim requires to "fuse" an atom with a bimodal wavefunction together into one that resembles the ground state wavefunction by moving the well left and right and changing its depth. What context in quantum computing requires the fusing of wavefunction of atoms together. Is it in general a hard problem in the building of quantum computers?

Answer 7

thx for agreeing to the speaker session and being flexible on the date! some misc questions/ topics that have come to mind over the months

• do you have any knowledge of or can comment on the background on the Martinis-Google arrangement/ merger? it seemed to be a "bolt from the blue" without a lot of behind-the-scenes info available. was wondering who initiated this. were Brin/ Page personally involved? had they expressed earlier interest in quantum computing? what caught their eye in particular about the lab? what is their thinking on allowing the lab to be independent vs acquired by Google? did the environment change somehow after the acquisition? has it been a smooth transition etc?

• what initially inspired you to get into physics/ quantum computing? did you have exposure/ ideas on it even as an undergraduate? do you have any particular heroes in the field, and why?

• what were particular challenges with your Phd you recall? any "lessons learned"? it appears you skipped the masters degree, what was your decision process there, and are you satisfied with that choice?

• quantum computing has a lot of sci fi aspects and probably public misconceptions. what are common misconceptions? what do you think are the short and long term prospects for QM computing? where does it fit with conventional computing approaches?

• there are many labs worldwide pursuing QM computing. could you contrast your own lab approach to them? how much do you & other lab staff follow other research? is anyone particularly on the wrong or right track in your opinion? is your lab focusing on all key aspects or is there some area thats being left to others? also theres a lot of media dedicated to DWave. whats your professional/ personal opinion of their approaches, can you comment? also it looks like your lab has no experiments on adiabatic computing, whats your take on this variant? has potential, or a dead end, etc?

• there are QM computing skeptics, one that has come to mind and been "debunked" by others is Dyakonov. some other skeptics argue that noise will not scale well in QM computing systems and will be a severe roadblock. whats your take on noise scaling in QM computing?

• Aaronson is a very prominent QM computing advocate/ writer offering near "play by play" commentary on his blog and elsewhere and has critiqued Dwave strongly and seems to highly favor the Martinis lab approaches. fair/ well informed/ accurate? whats your take on his statements/ attitude?

• youve been active in the hbar chatroom for many months. what drew you into the chat room? what do you regard as its strengths/ weaknesses? if you could change one thing about the site or the chat room, what would it be?

• from hbar transcripts, recently you got married (congratulations!). any comments on work/life balance in science/ physics? possible, challenging, difficult? does this have any affect on your scientific research; is your routine changed at all?

• you have quite a bit of experience in programming. how does this connect with your scientific research? any advice on programming abilities/ skills for scientists? whats your favorite language, how do you use it?

• you mentioned in the chat room some challenging issues with peer review in submitting publication(s). whats your take on the peer review process, does it mostly work for you? could it be improved somehow? whats the biggest challenge with it? etc

• many nations have substantial quantum computing science (govt funded) budgets, Australia and Europe come to mind (the latter just announced recently). do you have any opinion/ comment on whether a national science budget for QM computing makes sense in the US? is the private funding sufficient or could there be some major additional benefit with public funding?

• what advice would you have for the younger generation for getting into physics, studying it, advancing, etc? does it have a big future? has it overall been a great/ fulfilling choice for you? can you name any downsides/ obstacles/ admonitions?

Answer 8

Quantum computing has been around now for nearly forty years, and hasn't delivered anything. Meanwhile ordinary computing has advanced in leaps and bounds. So: is quantum computing just some two-bit jam-tomorrow hype that's never going to deliver anything? Is it going to quietly fade away like string theory, having consumed millions of publicly-funded man-hours, all for nought? Man hours that could have been productively spent on say optical computing, as per Taming Light at the Nanoscale?