Toward a Coherent US Government Strategy for QIS

In an upbeat  recent post, Spiros reported some encouraging news about quantum information science from the US National Science and Technology Council. Today I’ll chime in with some further perspective and background.

report-cover-2The Interagency Working Group on Quantum Information Science (IWG on QIS), which began its work in late 2014, was charged “to assess Federal programs in QIS, monitor the state of the field, provide a forum for interagency coordination and collaboration, and engage in strategic planning of Federal QIS activities and investments.”  The IWG recently released a  well-crafted report, Advancing Quantum Information Science: National Challenges and Opportunities. The report recommends that “quantum information science be considered a priority for Federal coordination and investment.”

All the major US government agencies supporting QIS were represented on the IWG, which was co-chaired by officials from DOE, NSF, and NIST:

  • Steve Binkley, who heads the Advanced Scientific Computing Research (ASCR) program in the Department of Energy Office of Science,
  • Denise Caldwell, who directs the Physics Division of the National Science Foundation,
  • Carl Williams, Deputy Director of the Physical Measurement Laboratory at the National Institute for Standards and Technology.

Denise and Carl have been effective supporters of QIS over many years of government service. Steve has recently emerged as another eloquent advocate for the field’s promise and importance.

At our request, the three co-chairs fielded questions about the report, with the understanding that their responses would be broadly disseminated. Their comments reinforced the message of the report — that all cognizant agencies favor a “coherent, all-of-government approach to QIS.”

Science funding in the US differs from elsewhere in the world. QIS is a prime example — for over 20 years, various US government agencies, each with its own mission, goals, and culture, have had a stake in QIS research. By providing more options for supporting innovative ideas, the existence of diverse autonomous funding agencies can be a blessing. But it can also be bewildering for scientists seeking support, and it poses challenges for formulating and executing effective national science policy. It’s significant that many different agencies worked together in the IWG, and were able to align with a shared vision.

“I think that everybody in the group has the same goals,” Denise told us. “The nation has a tremendous opportunity here. This is a terrifically important field for all of us involved, and we all want to see it succeed.” Carl added, “All of us believe that this is an area in which the US must be competitive, it is very important for both scientific and technological reasons … The differences [among agencies] are minor.”

Asked about the timing of the IWG and its report, Carl noted the recent trend toward “emerging niche applications” of QIS such as quantum sensors, and Denise remarked that government agencies are responding to a plea from industry for a cross-disciplinary work force broadly trained in QIS. At the same time, Denise emphasized, the IWG recognizes that “there are still many open basic science questions that are important for this field, and we need to focus investment onto these basic science questions, as well as look at investments or opportunities that lead into the first applications.”

DOE’s FY2017 budget request includes $10M to fund a new QIS research program, coordinated with NIST and NSF. Steve explained the thinking behind that request:  “There are problems in the physical science space, spanned by DOE Office of Science programs, where quantum computation would be a useful a tool. This is the time to start making investments in that area.” Asked about the longer term commitment of DOE to QIS research, Steve was cautious. “What it will grow into over time is hard to tell — we’re right at the beginning.”

What can the rest of us in the QIS community do to amplify the impact of the report? Carl advised: “All of us should continue getting the excitement of the field out there, [and point to] the potential long term payoffs,  whether they be in searches for dark matter or building better clocks or better GPS systems or better sensors. Making everybody aware of all the potential is good for our economy, for our country, and for all of us.”

Taking an even longer view, Denise reminded us that effective advocacy for QIS can get young people “excited about a field they can work in, where they can get jobs, where they can pursue science — that can be critically important.  If we all think back to our own beginning careers, at some point in time we got excited about science. And so whatever one can do to excite the next generation about science and technology, with the hope of bringing them into studying and developing careers in this field, to me this is tremendously valuable. ”

All of us in the quantum information science community owe a debt to the IWG for their hard work and eloquent report, and to the agencies they represent for their vision and support. And we are all fortunate to be participating in the early stages of a new quantum revolution. As the IWG report makes clear, the best is yet to come.

Quantum Supremacy: The US gets serious

If you have been paying any attention to the news on quantum computing and the evolution of industrial and national efforts towards realizing a scalable, fault-tolerant quantum computer that can tackle problems intractable to current supercomputing capabilities, then you know that something big is stirring throughout the quantum world.

More than 15 years ago, Microsoft decided to jump into the quantum computing business betting big on topological quantum computing as the next big thing. The new website of Microsoft’s Station Q shows that keeping a low profile is no longer an option. This is a sentiment that Google clearly shared, when back in 2013, they decided to promote their new partnership with NASA Ames and D-Wave, known as the Quantum A.I. Lab, through a YouTube video that went viral (disclosure: they do own Youtube.) In fact, IQIM worked with Google at the time to get kids excited about the quantum world by developing qCraft, a mod introducing quantum physics into the world of Minecraft. Then, a few months ago, IBM unveiled the quantum experience website, which captured the public’s imagination by offering a do-it-yourself opportunity to run an algorithm on a 5-qubit quantum chip in the cloud.

But, looking at the opportunities for investment in academic groups working on quantum computing, companies like Microsoft were/are investing heavily in experimental labs across the pond, such as Leo Kowenhoven’s group at TU Delft and Charlie Marcus’ group in Copenhagen, with smaller investments here in the US. This may just reflect the fact that the best efforts to build topological qubits are in Europe, but it still begs the question why a fantastic idea like topologically protected majorana zero modes, by Yale University’s Nick Read and Dmitry Green, which inspired the now famous Majorana wire paper by Alexei Kitaev when he was a researcher at Microsoft’s Redmond research lab, and whose transition from theory to experiment took off with contributions from Maryland and IQIM researchers, was outsourced to European labs for experimental verification and further development. The one example of a large investment in a US academic research group has been Google’s hiring of John Martinis away from UCSB. In fact, John and I met a couple of years ago to discuss investment into his superconducting quantum computing efforts, because government funding for academic efforts to actually build a quantum computer was lacking. China was investing, Canada was investing, Europe went a little crazy, but the US was relying on visionary agencies like IARPA, DARPA and the NSF to foot the bill (without which Physics Frontiers Centers like IQIM wouldn’t be around). In short, there was no top-down policy directive to focus national attention and inter-agency Federal funding on winning the quantum supremacy race.

Until now.

The National Science and Technology Council, which is chaired by the President of the United States and “is the principal means within the executive branch to coordinate science and technology policy across the diverse entities that make up the Federal research and development enterprise”, just released the following report:

Advancing Quantum Information Science: National Challenges and Opportunities

The White House blog post does a good job at describing the high-level view of what the report is about and what the policy recommendations are. There is mention of quantum sensors and metrology, of the promise of quantum computing to material science and basic science, and they even go into the exciting connections between quantum error-correcting codes and emergent spacetime, by IQIM’s Pastawski, et al.

But the big news is that the report recommends significant and sustained investment in Quantum Information Science. The blog post reports that the administration intends “to engage academia, industry, and government in the upcoming months to … exchange views on key needs and opportunities, and consider how to maintain vibrant and robust national ecosystems for QIS research and development and for high-performance computing.”

Personally, I am excited to see how the fierce competition at the academic, industrial and now international level will lead to a race for quantum supremacy. The rivals are all worthy of respect, especially because they are vying for supremacy not just over each other, but over a problem so big and so interesting, that anyone’s success is everyone’s success. After all, anyone can quantum, and if things go according to plan, we will soon have the first generation of kids trained on (it doesn’t exist yet), as well as Until then, quantum chess and qCraft will have to do.

LIGO: Playing the long game, and winning big!

Wow. What a day! And what a story!

Kip Thorne in 1972, around the time MTW was completed.

Kip Thorne in 1972, around the time MTW was completed.

It is hard for me to believe, but I have been on the Caltech faculty for nearly a third of a century. And when I arrived in 1983, interferometric detection of gravitational waves was already a hot topic of discussion here. At Kip Thorne’s urging, Ron Drever had been recruited to Caltech and was building the 40-meter prototype interferometer (which is still operating as a testbed for future detection technologies). Kip and his colleagues, spurred by Vladimir Braginsky’s insights, had for several years been actively studying the fundamental limits of quantum measurement precision, and how these might impact the search for gravitational waves.

I decided to bone up a bit on the subject, so naturally I pulled down from my shelf the “telephone book” — Misner, Thorne, and Wheeler’s mammoth Gravitationand browsed Chapter 37 (Detection of Gravitational Wave), for which Kip had been the lead author. The chapter brimmed over with enthusiasm for the subject, but to my surprise interferometers were hardly mentioned. Instead the emphasis was on mechanical bar detectors. These had been pioneered by Joseph Weber, whose efforts in the 1960s had first aroused Kip’s interest in detecting gravitational waves, and by Braginsky.

I sought Kip out for an explanation, and with characteristic clarity and patience he told how his views had evolved. He had realized in the 1970s that a strain sensitivity of order 10^{-21} would be needed for a good chance at detection, and after many discussions with colleagues like Drever, Braginsky, and Rai Weiss, he had decided that kind of sensitivity would not be achievable with foreseeable technology using bars.

Ron Drever, who built Caltech's 40-meter prototype interferometer in the 1980s.

Ron Drever, who built Caltech’s 40-meter prototype interferometer in the 1980s.

We talked about what would be needed — a kilometer scale detector capable of sensing displacements of 10^{-18} meters. I laughed. As he had many times by then, Kip told why this goal was not completely crazy, if there is enough light in an interferometer, which bounces back and forth many times as a waveform passes. Immediately after the discussion ended I went to my desk and did some crude calculations. The numbers kind of worked, but I shook my head, unconvinced. This was going to be a huge undertaking. Success seemed unlikely. Poor Kip!

I’ve never been involved in LIGO, but Kip and I remained friends, and every now and then he would give me the inside scoop on the latest developments (most memorably while walking the streets of London for hours on a beautiful spring evening in 1991). From afar I followed the forced partnership between Caltech and MIT that was forged in the 1980s, and the painful transition from a small project under the leadership of Drever-Thorne-Weiss (great scientists but lacking much needed management expertise) to a large collaboration under a succession of strong leaders, all based at Caltech.

Vladimir Braginsky, who realized that quantum effects constrain gravitational wave detectors.

Vladimir Braginsky, who realized that quantum effects limit the sensitivity of  gravitational wave detectors.

During 1994-95, I co-chaired a committee formulating a long-range plan for Caltech physics, and we spent more time talking about LIGO than any other issue. Part of our concern was whether a small institution like Caltech could absorb such a large project, which was growing explosively and straining Institute resources. And we also worried about whether LIGO would ultimately succeed. But our biggest worry of all was different — could Caltech remain at the forefront of gravitational wave research so that if and when LIGO hit paydirt we would reap the scientific benefits?

A lot has changed since then. After searching for years we made two crucial new faculty appointments: theorist Yanbei Chen (2007), who provided seminal ideas for improving sensitivity, and experimentalist Rana Adhikari (2006), a magician at the black art of making an interferometer really work. Alan Weinstein transitioned from high energy physics to become a leader of LIGO data analysis. We established a world-class numerical relativity group, now led by Mark Scheel. Staff scientists like Stan Whitcomb also had an essential role, as did longtime Project Manager Gary Sanders. LIGO Directors Robbie Vogt, Barry Barish, Jay Marx, and now Dave Reitze have provided effective and much needed leadership.

Rai Weiss, around the time he conceived LIGO in an amazing 1972 paper.

Rai Weiss, around the time he conceived LIGO in an amazing 1972 paper.

My closest connection to LIGO arose during the 1998-99 academic year, when Kip asked me to participate in a “QND reading group” he organized. (QND stands for Quantum Non-Demolition, Braginsky’s term for measurements that surpass the naïve quantum limits on measurement precision.) At that time we envisioned that Advanced LIGO would turn on in 2008, yet there were still many questions about how it would achieve the sensitivity required to ensure detection. I took part enthusiastically, and learned a lot, but never contributed any ideas of enduring value. The discussions that year did have positive outcomes, however; leading for example to a seminal paper by Kimble, Levin, Matsko, Thorne, and Vyatchanin on improving precision through squeezing of light. By the end of the year I had gained a much better appreciation of the strength of the LIGO team, and had accepted that Advanced LIGO might actually work!

I once asked Vladimir Braginsky why he spent years working on bar detectors for gravitational waves, while at the same time realizing that fundamental limits on quantum measurement would make successful detection very unlikely. Why wasn’t he trying to build an interferometer already in the 1970s? Braginsky loved to be asked questions like this, and his answer was a long story, told with many dramatic flourishes. The short answer is that he viewed interferometric detection of gravitational waves as too ambitious. A bar detector was something he could build in his lab, while an interferometer of the appropriate scale would be a long-term project involving a much larger, technically diverse team.

Joe Weber, who audaciously believed gravitational waves can be detected on earth.

Joe Weber, whose audacious belief that gravitational waves are detectable on earth inspired Kip Thorne and many others.

Kip’s chapter in MTW ends with section 37.10 (“Looking toward the future”) which concludes with this juicy quote (written almost 45 years ago):

“The technical difficulties to be surmounted in constructing such detectors are enormous. But physicists are ingenious; and with the impetus provided by Joseph Weber’s pioneering work, and with the support of a broad lay public sincerely interested in pioneering in science, all obstacles will surely be overcome.”

That’s what we call vision, folks. You might also call it cockeyed optimism, but without optimism great things would never happen.

Optimism alone is not enough. For something like the detection of gravitational waves, we needed technical ingenuity, wise leadership, lots and lots of persistence, the will to overcome adversity, and ultimately the efforts of hundreds of hard working, talented scientists and engineers. Not to mention the courage displayed by the National Science Foundation in supporting such a risky project for decades.

I have never been prouder than I am today to be part of the Caltech family.

Wouldn’t you like to know what’s going on in my mind?

I suppose most theoretical physicists who (like me) are comfortably past the age of 60 worry about their susceptibility to “crazy-old-guy syndrome.” (Sorry for the sexism, but all the victims of this malady I know are guys.) It can be sad when a formerly great scientist falls far out of the mainstream and seems to be spouting nonsense.

Matthew Fisher is only 55, but reluctance to be seen as a crazy old guy might partially explain why he has kept pretty quiet about his passionate pursuit of neuroscience over the past three years. That changed two months ago when he posted a paper on the arXiv about Quantum Cognition.

Neuroscience has a very seductive pull, because it is at once very accessible and very inaccessible. While a theoretical physicist might think and write about a brane even without having or seeing a brane, everybody’s got a brain (some scarecrows excepted). On the other hand, while it’s not too hard to write down and study the equations that describe a brane, it is not at all easy to write down the equations for a brain, let alone solve them. The brain is fascinating because we know so little about it. And … how can anyone with a healthy appreciation for Gödel’s Theorem not be intrigued by the very idea of a brain that thinks about itself?

(Almost) everybody's got a brain.

(Almost) everybody’s got a brain.

The idea that quantum effects could have an important role in brain function is not new, but is routinely dismissed as wildly implausible. Matthew Fisher begs to differ. And those who read his paper (as I hope many will) are bound to conclude: This old guy’s not so crazy. He may be onto something. At least he’s raising some very interesting questions.

My appreciation for Matthew and his paper was heightened further this Wednesday, when Matthew stopped by Caltech for a lunch-time seminar and one of my interminable dinner-time group meetings. I don’t know whether my brain is performing quantum information processing (and neither does Matthew), but just the thought that it might be is lighting me up like a zebrafish.

Following Matthew, let’s take a deep breath and ask ourselves: What would need to be true for quantum information processing to be important in the brain? Presumably we would need ways to (1) store quantum information for a long time, (2) transport quantum information, (3) create entanglement, and (4) have entanglement influence the firing of neurons. After a three-year quest, Matthew has interesting things to say about all of these issues. For details, you should read the paper.

Matthew argues that the only plausible repositories for quantum information in the brain are the Phosphorus-31 nuclear spins in phosphate ions. Because these nuclei are spin-1/2, they have no electric quadrupole moments and hence corresponding long coherence times — of order a second. That may not be long enough, but phosphate ions can be bound with calcium ions into objects called Posner clusters, each containing six P-31 nuclei. The phosphorus nuclei in Posner clusters might have coherence times greatly enhanced by motional narrowing, perhaps as long as weeks or even longer.

Where energy is being consumed in a cell, ATP sometimes releases diphosphate ions (what biochemists call pyrophosphate), which are later broken into two separate phosphate ions, each with a single P-31 qubit. Matthew argues that the breakup of the diphosphate, catalyzed by a suitable enzyme, will occur at an enhanced rate when these two P-31 qubits are in a spin singlet rather than a spin triplet. The reason is that the enzyme has to grab ahold of the diphosphate molecule and stop its rotation in order to break it apart, which is much easier when the molecule has even rather than odd orbital angular momentum; therefore due to Fermi statistics the spin state of the P-31 nuclei must be antisymmetric. Thus wherever ATP is consumed there is a plentiful source of entangled qubit pairs.

If the phosphate molecules remain unbound, this entanglement will decay in about a second, but it is a different story if the phosphate ions group together quickly enough into Posner clusters, allowing the entanglement to survive for a much longer time. If the two members of an entangled qubit pair are snatched up by different Posner clusters, the clusters may then be transported into different cells, distributing the entanglement over relatively long distances.

(a) Two entangled Posner clusters. Each dot is a P-31 nuclear spin, and each dashed line represents a singlet pair. (b) Many entangled Posner clusters. [From the paper]

(a) Two entangled Posner clusters. Each dot is a P-31 nuclear spin, and each dashed line represents a singlet pair. (b) Many entangled Posner clusters. [From Fisher 2015]

What causes a neuron to fire is a complicated story that I won’t attempt to wade into. Suffice it to say that part of the story may involve the chemical binding of a pair of Posner clusters which then melt if the environment is sufficiently acidic, releasing calcium ions and phosphate ions which enhance the firing. The melting rate depends on the spin state of the six P-31 nuclei within the cluster, so that entanglement between clusters in different cells may induce nonlocal correlations among different neurons, which could be quite complex if entanglement is widely distributed.

This scenario raises more questions than it answers, but these are definitely scientific questions inviting further investigation and experimental exploration. One thing that is far from clear at this stage is whether such quantum correlations among neurons (if they exist at all) would be easy to simulate with a classical computer. Even if that turns out to be so, these potential quantum effects involving many neurons could be fabulously interesting. IQIM’s mission is to reach for transformative quantum science, particularly approaches that take advantage of synergies between different fields of study. This topic certainly qualifies.* It’s going to be great fun to see where it leads.

If you are a young and ambitious scientist, you may be contemplating the dilemma: Should I pursue quantum physics or neuroscience? Maybe, just maybe, the right answer is: Both.

*Matthew is the only member of the IQIM faculty who is not a Caltech professor, though he once was.

Kitaev, Moore, Read share Dirac Medal!

Since its founding 30 years ago, the Dirac Medal has been one of the most prestigious honors in theoretical physics. Particle theorists and string theorists have claimed most of the medals, but occasionally other fields break through, as when Haldane, Kane, and Zhang shared the 2012 Dirac Medal for their pioneering work on topological insulators. I was excited to learn today that the 2015 Dirac Medal has been awarded to Alexei Kitaev, Greg Moore, and Nick Read “for their interdisciplinary contributions which introduced  concepts of conformal field theory and non-abelian quasiparticle statistics in condensed matter systems and  applications of these ideas to quantum computation.”

Left to right: Alexei Kitaev, Greg Moore and Nicholas Read.

Left to right: Alexei Kitaev, Greg Moore, and Nick Read.

I have written before about the exciting day in April 1997 when Alesha and I met, and I heard for the first time about the thrilling concept of a topological quantum computer. I’ll take the liberty of drawing a quote from that post, which seems particularly relevant today:

Over coffee at the Red Door Cafe that afternoon, we bonded over our shared admiration for a visionary paper by Greg Moore and Nick Read about non-abelian anyons in fractional quantum Hall systems, though neither of us fully understood the paper (and I still don’t). Maybe, we mused together, non-abelian anyons are not just a theorist’s dream … It was the beginning of a beautiful friendship.

As all physics students know, fundamental particles in three spatial dimensions come in two varieties, bosons and fermions, but in two spatial dimensions more exotic possibilities abound, dubbed “anyons” by Wilczek. Anyons have an exotic spin, a fraction of an electron’s spin, and corresponding exotic statistics — when one anyon is carried around another, their quantum state picks up a nontrivial topological phase. (I had some fun discussions with Frank Wilczek in 1981 as he was developing the theory of anyons. In some of his writings Frank has kindly credited me for suggesting to him that a robust spin-statistics connection should hold in two dimensions, so that fractional spin is necessarily accompanied by fractional statistics. The truth is that my understanding of this point was murky at best back then.) Not long after Wilczek’s paper, Bert Halperin recognized the relevance of anyons to the strange fractional quantum Hall states that had recently been discovered; these support particle-like objects carrying a fraction of the electron’s electric charge, which Halperin recognized to be anyons.

Non-abelian anyons are even more exotic. In a system with many widely separated non-abelian anyons, there are a vast number of different ways for the particles to “fuse” together, giving rise to many possible quantum states, all of which are in principle distinguishable but in practice are hard to tell apart. Furthermore, by “braiding” the anyons (performing a sequence of particle exchanges, so the world lines of the anyons trace out a braid in three-dimensional spacetime), this state can be manipulated, coherently processing the quantum information encoded in the system.

Others (including me) had mused about non-abelian anyons before Moore and Read came along, but no one had proposed a plausible story for how such exotic objects would arise in a realistic laboratory setting. As collaborators, Moore and Read complemented one another perfectly. Greg was, and is, one of the world’s leading experts on conformal field theory. Nick was, and is, one of the world’s leading experts on the fractional quantum Hall effect. Together, they realized that one of the already known fractional quantum Hall states (at filling factor 5/2) is a good candidate for a topological phase supporting non-abelian anyons. This was an inspired guess, most likely correct, though we still don’t have smoking gun experimental evidence 25 years later. Their paper is a magical and rare combination of mathematical sophistication with brilliant intuition.

Alexei arrived at his ideas about non-abelian anyons coming from a different direction, though I suspect he drew inspiration from the earlier deep contributions of Moore and Read. He was trying to imagine a physical system that could store and process a quantum state reliably. Normally quantum systems are very fragile — just looking at the system alters its state. To prevent a quantum computer from making errors, we need to isolate the information processed by the computer from the environment. A system of non-abelian anyons has just the right properties to make this possible; it carries lots of information, but the environment can’t read (or damage) that information when it looks at the particles one at a time. That’s because the information is not encoded in the individual particles, but instead in subtle collective properties shared by many particles at once.

Alexei and I had inspiring discussions about topological quantum computing when we first met at Caltech in April 1997, which continued at a meeting in Torino, Italy that summer, where we shared a bedroom. I was usually asleep by the time he came to bed, because he was staying up late, typing his paper.

Alexei did not think it important to publish his now renowned 1997 paper in a journal — he was content for the paper to be accessible on the arXiv. But after a few years I started to get worried … in my eyes Alexei was becoming an increasingly likely Nobel Prize candidate. Would it cause a problem if his most famous paper had never been published? Just to be safe, I arranged for it to appear in Annals of Physics in 2003, where I was on the editorial board at the time. Frank Wilczek, then the editor, was delighted by this submission, which has definitely boosted the journal’s impact factor! (“Fault-tolerant quantum computation by anyons” has 2633 citations as of today, according to Google Scholar.) Nobelists are ineligible for the Dirac Medal, but some past medalists have proceeded to greater glory. It could happen again, right?

Alesha and I have now been close friends and collaborators for 18 years, but I have actually known Greg and Nick even longer. I taught at Harvard for a few years in the early 1980s, at a time when an amazingly talented crew of physics graduate students roamed the halls, of whom Andy Cohen, Jacques Distler, Ben Grinstein, David Kaplan, Aneesh Manohar, Ann Nelson, and Phil Nelson among others all made indelible impressions. But there was something special about Greg. The word that comes to mind is intensity. Few students exhibit as much drive and passion for physics as Greg did in those days. He’s calmer now, but still pretty intense. I met Nick a few years later when we tried to recruit him to the Caltech faculty. Luring him to southern California turned out to be a lost cause because he didn’t know how to drive a car. I suppose he’s learned by now?* Whenever I’ve spoken to Nick in the years since then, I’ve always been dazzled by his clarity of thought.

Non-abelian anyons are at a pivotal stage, with lots of experimental hints supporting their existence, but still no ironclad evidence. I feel confident this will change in the next few years. These are exciting times!

And guess what? This occasion gives me another opportunity to dust off one of my poems!

Anyon, Anyon

Anyon, anyon, where do you roam?
Braid for a while before you go home.

Though you’re condemned just to slide on a table,
A life in 2D also means that you’re able
To be of a type neither Fermi nor Bose
And to know left from right — that’s a kick, I suppose.

You and your buddy were made in a pair
Then wandered around, braiding here, braiding there.
You’ll fuse back together when braiding is through
We’ll bid you adieu as you vanish from view.

Alexei exhibits a knack for persuading
That someday we’ll crunch quantum data by braiding,
With quantum states hidden where no one can see,
Protected from damage through top-ology.

Anyon, anyon, where do you roam?
Braid for a while, before you go home.

*Note added: Nick confirms, “Yes, I’ve had a driving license since 1992, and a car since 1994!”

A detective with a quantum helper

Have you ever wanted to be incredibly perceptive and make far-reaching deductions about people? I have always been fascinated by spy stories, and how the main character in them notices tiny details of his surroundings to navigate life-or-death situations. This skill seems out of reach for us normal people; you have to be “a high-functioning sociopath” to memorize all existing data on behavior, clothes choices and forensic science. Of course I’m referring to:
Small details help Sherlock figure out what did the woman do to meet such a sad end
Yet in the not too distant future, a computer may help you become a brilliant detective (or a scheming villain) yourself! The first step is noticing the details, which is known in machine learning as the classification task. Here is a pioneering work that somewhat resembles the above picture, only it’s done by a computer:
A computer spits out a sentence (read down) describing what's in the picture. Work by Stanford group.

The task for the computer here was to produce a verbal description of the image. There are thousands of words in the vocabulary, and a computer has to try them in different combinations to make a sensible sentence. There is no way a computer can be given an exhaustive list of correct sentences with examples of images for each. That kind of list would be a database bigger than the earth (as one can see just by counting the number of combinations). So to train the computer to use language like in a picture above, one only possesses a limited set of examples – maybe a few thousand pictures with descriptions. Yet we as humans are capable of learning from just seeing a few examples, by noticing the repeating patterns. So the computer can do the same! The score next to each word above is an estimate based on those few thousand examples of how relevant is the word “tennis” or “woman” to what’s in the box on the image. The algorithm produces possible sentences, scores them, and then selects the sentence with the highest total score.

Once the classification task is done, one needs to use all the collected information to make a prediction – as Sherlock is able to point out the most probable motive in the first picture, we also want to predict a piece of very personal information: we’d like to know how to start up a conversation with that tennis player.

Humans are actually good at classification tasks: with luck, we can notice and type in our cellphone all the details the predictor will need, like brand of clothing, hair color, height… though computers recently became better than humans at facial expression recognition, so we don’t have to trust ourselves on that anymore. Finally, when all the data is collected, most humans will still say only generic advice to you on conversation starters. Which means we are very bad at prediction tasks. We don’t notice the hidden dependencies between brand of clothes and sense of humor. But such information may not hide from the all-seeing eye of the machine learning algorithm! So expect your cellphones to give you dating advice within 10 years… 

Now how do quantum computers come into play? Well if you look at your search results, they are still pretty irrelevant most of the time. Imagine you used them as conversation starters – you’ll embarrass yourself 9 out of 10 times! To make this better, a certain company needs more memory and processing power. Yet most advanced deep learning routines remain out of reach, just because there are exponentially many hidden dependencies one would need to try and reject before the algorithm finds the right predictor. So a certain company turns to us, quantum computing people, as we deal with exponentially hard problems notoriously well! And indeed, quantum algorithms make some of the machine learning routines exponentially faster – see this Quantum Machine Learning article, as well as a talk by Seth Lloyd for technical details. Some anonymous stock trader is already trying to intimidate their fellow quants (quantitative analysts) by calling the top trading system “Quantum machine learning”. I think we should appreciate his sense of humor and invest into his algorithm as soon as opens such functionality. Or we could invest in Teagan from Caltech – her code recently won the futures contest on the same website.

When I met with Steven Spielberg to talk about Interstellar

Today I had the awesome and eagerly anticipated privilege of attending a screening of the new film Interstellar, directed by Christopher Nolan. One can’t help but be impressed by Nolan’s fertile visual imagination. But you should know that Caltech’s own Kip Thorne also had a vital role in this project. Indeed, were there no Kip Thorne, Interstellar would never have happened.

On June 2, 2006, I participated in an unusual one-day meeting at Caltech, organized by Kip and the movie producer Lynda Obst (Sleepless in Seattle, Contact, The Invention of Lying, …). Lynda and Kip, who have been close since being introduced by their mutual friend Carl Sagan decades ago, had conceived a movie project together, and had collaborated on a “treatment” outlining the story idea. The treatment adhered to a core principle that was very important to Kip — that the movie be scientifically accurate. Though the story indulged in some wild speculations, at Kip’s insistence it skirted away from any flagrant violation of the firmly established laws of Nature. This principle of scientifically constrained speculation intrigued Steven Spielberg, who was interested in directing.

The purpose of the meeting was to brainstorm about the story and the science behind it with Spielberg, Obst, and Thorne. A remarkable group assembled, including physicists (Andrei Linde, Lisa Randall, Savas Dimopoulos, Mark Wise, as well as Kip), astrobiologists (Frank Drake, David Grinspoon), planetary scientists (Alan Boss, John Spencer, Dave Stevenson), and psychologists (Jay Buckey, James Carter, David Musson). As we all chatted and got acquainted, I couldn’t help but feel that we were taking part in the opening scene of a movie about making a movie. Spielberg came late and left early, but spent about three hours with us; he even brought along his Dad (an engineer).

Time_cover_interstellarThough the official release of Interstellar is still a few days away, you may already know from numerous media reports (including the cover story in this week’s Time Magazine) the essential elements of the story, which involves traveling through a wormhole seeking a new planet for humankind, a replacement for the hopelessly ravaged earth. The narrative evolved substantially as the project progressed, but traveling through a wormhole to visit a distant planet was already central to the original story.

Inevitably, some elements of the Obst/Thorne treatment did not survive in the final film. For one, Stephen Hawking was a prominent character in the original story; he joined the mission because of his unparalleled expertise at wormhole transversal, and Stephen’s ALS symptoms eased during prolonged weightlessness, only to recur upon return to earth gravity. Also, gravitational waves played a big part in the treatment; in particular the opening scene depicted LIGO scientists discovering the wormhole by detecting the gravitational waves emanating from it.

There was plenty to discuss to fill our one-day workshop, including: the rocket technology needed for the trip, the strong but stretchy materials that would allow the ship to pass through the wormhole without being torn apart by tidal gravity, how to select a crew psychologically fit for such a dangerous mission, what exotic life forms might be found on other worlds, how to communicate with an advanced civilization which resides in a higher dimensional bulk rather than the three-dimensional brane to which we’re confined, how to build a wormhole that stays open rather than pinching off and crushing those who attempt to pass through, and whether a wormhole could enable travel backward in time.

Spielberg was quite engaged in our discussions. Upon his arrival I immediately shot off a text to my daughter Carina: “Steven Spielberg is wearing a Brown University cap!” (Carina was a Brown student at the time, as Spielberg’s daughter had been.) Steven assured us of his keen interest in the project, noting wryly that “Aliens have been very good to me,” and he mentioned some of his favorite space movies, which included some I had also enjoyed as a kid, like Forbidden Planet and (the original) The Day the Earth Stood Still. In one notable moment, Spielberg asked the group “Who believes that intelligent life exists elsewhere in the universe?” We all raised our hands. “And who believes that the earth has been visited by extraterrestrial civilizations?” No one raised a hand. Steven seemed struck by our unanimity, on both questions.

I remember tentatively suggesting that the extraterrestrials had mastered M-theory, thus attaining computational power far beyond the comprehension of earthlings, and that they themselves were really advanced robots, constructed by an earlier generation of computers. Like many of the fun story ideas floated that day, this one had no apparent impact on the final version of the film.

Spielberg later brought in Jonah Nolan to write the screenplay. When Spielberg had to abandon the project because his DreamWorks production company broke up with Paramount Pictures (which owned the story), Jonah’s brother Chris Nolan eventually took over the project. Jonah and Chris Nolan transformed the story, but continued to consult extensively with Kip, who became an Executive Producer and says he is pleased with the final result.

Of the many recent articles about Interstellar, one of the most interesting is this one in Wired by Adam Rogers, which describes how Kip worked closely with the visual effects team at Double Negative to ensure that wormholes and rapidly rotating black holes are accurately depicted in the film (though liberties were taken to avoid confusing the audience). The images produced by sophisticated ray tracing computations were so surprising that at first Kip thought there must be a bug in the software, though eventually he accepted that the calculations are correct, and he is still working hard to more fully understand the results.

ScienceofInterstellarMech.inddI can’t give away the ending of the movie, but I can safely say this: When it’s over you’re going to have a lot of questions. Fortunately for all of us, Kip’s book The Science of Interstellar will be available the same day the movie goes into wide release (November 7), so we’ll all know where to seek enlightenment.

In fact on that very same day we’ll be treated to the release of The Theory of Everything, a biopic about Stephen and Jane Hawking. So November 7 is going to be an unforgettable Black Hole Day. Enjoy!