# A Discussion and a Debate

## Heavier than air flight of the 21 century?

The very first post on this blog entitled “Combinatorics, Mathematics, Academics, Polemics, …” asked the question “Are mathematical debates possible?” We also had posts devoted to debates and to controversies.

A few days ago, the first post in a discussion between Aram Harrow, a brilliant computer scientists and quantum information researcher (and a decorated debator), and myself on quantum error correction was launched in Dick Lipton and Ken Regan’s big-league blog, Gödel’s Lost Letter and P=NP.

The central question we would like to discuss is:

Are universal quantum computers based on quantum error correction possible.

In preparation for the public posts, Ken, Aram, Dick, and me are having very fruitful and interesting email discussion on this and related matters, and also the post itself have already led to very interesting comments. Ken is doing marvels in editing what we write.

Dick opened the post by talking on perpetual motion machines which is ingenious because it challenges both sides of the discussion. Perpetual motion turned out to be impossible: will quantum computers enjoy the same fate? On the other hand (and closer to the issue at hand), an argument against quantum mechanics based on the impossibility of perpetual motion by no other than Einstein turned out to be false, are skeptical ideas to quantum computers just as bogus? (The answer could be yes to both questions.) Some people claimed that heavier-than-air flight might is a better analogy. Sure, perhaps it is better.

But, of course, analogies with many human endeavors can be made, and for these endeavors, some went one way, and some went the other way, and for some we don’t know.

Although this event is declared as a debate, I would like to think about it as a discussion. In the time scale of such a discussion what we can hope for is to better understand each other positions, and, not less important, to better understand our own positions.  (Maybe I will comment here about some meta aspects of this developing discussion/debate.)

## A real debate

A real emerging debate is if we (scientists) should boycott Elsevier. I tend to be against such an action, and especially against including refereeing papers for journals published by Elsevier as part of the boycott. I liked, generally speaking,  Gowers’s critical post on Elsevier, but the winds of war and associated rhetoric are not to my liking.  The universities are quite powerful, and they deal, negotiate and struggle with scientific publishers, and other similar bodies, on a regular  basis. I tend to think that the community of scientists should not be part of such struggles and that such involvement will harm the community and science. This is a real debate! But it looks almost over.  Many scientists joined the boycott and not many opposing opinions were made. It looks that we will have a little war and see some action. Exciting, as ever.

# Aaronson and Arkhipov’s Result on Hierarchy Collapse

Scott Aaronson gave a thought-provoking lecture in our Theory seminar three weeks ago.  (Actually, this was eleven months ago.) The slides are here . The lecture discussed two results regarding the computational power of quantum computers. One result from this paper gives an oracle-evidence that there are problems in BQP outside the polynomial hierarchy.  The method is based on “magnification” of results on bounded depth circuits. (It is related to the Linial-Nisan conjecture.)

The second result that we are going to discuss in this post (along with some of my provoked thoughts) is a recent result of Scott Aaronson and Alex Arkhipov which asserts that if  the power of quantum computers can be simulated by digital computers  then the polynomial hierarchy collapses.  More precisely, their result asserts that if sampling  probability distribution created by quantum computers (this is abbreviated as QSAMPLE) is in P then the polynomial hieararchy collapses to its third level.

The beautiful and important paper of Aaronson and Arkhipov is now out. Its main point of view is related to what I describe in the sixth section about “photon machines”. Update: Let me mention related idependent results by Michael J. Bremner, Richard Jozsa, Dan J. Shepherd in the paper entitled: “Classical simulation of commuting quantum computations implies collapse of the polynomial hierarchy“.

Here is the plan for this post

1) The polynomial hierarchy and results about hierarchy collapse

2) The Aaronson Arkhipov (AA) theorem and its proof

3) Two problems posed by AA

4) Does fault tolerance allow QSAMPLE (on the nose)? (Answer: yes)

5) Is there a quantum analog to Aaronson and Arkhipov’s result and what is the computational power of quantum computers?

6) Three Two competing scenarios

7) Aaronson and Arkhipov photon machines and the complexity class BOSONSAMPLING.

# Octonions to the Rescue

## Xavier Dahan and Jean-Pierre Tillich’s Octonion-based Ramanujan Graphs with High Girth.

Update (February 2012): Non associative computations can be trickier than we expect. Unfortunately, the paper by Dahan and Tillich turned out to be incorrect.

Update: There is more to be told about the background of the new exciting paper. In particular, I would like to tell you more about regular graphs with high girth. (I started below.) The Ramanujan graphs story is, of course, also fascinating so at the very least I should give good links.

Michael Atiyah‘s lecture at IAS physics last Friday was entertaining, educational and quite provocative.

The talk started with the following thesis: There are four fundamental forces of nature and there are four division rings over the reals. The real numbers, complex numbers, Quaternions and the Octonions. Atiyah expects that the Octonions will play a major role in physics and will allow a theory which accounts for gravitation. He described some specific steps in this direction and related ideas and connections. At the end of the talk,  Atiyah’s thesis looked more plausible than in the beginning. His concluding line was: “you can regard what I say as nonsense, or you can claim that you know it already, but you cannot make these two claims together.” In any case, it looks that the people in the audience were rather impressed by and sympathetic to the Octonionic ideas of this wise energetic scientific tycoon.

The same day I received an email from Nati Linial. The subject was: “a good topic for your blog” and the email contained just a single link.
http://arxiv.org/PS_cache/arxiv/pdf/1011/1011.2642v1.pdf

Nati is my older academic brother and often I regard our relations as similar to typical relations between older and younger (biological) brothers. When he tells me what to do I often rebel, but usually at the end I do as he says and most of the times he is right.

So I waited a couple of hours before looking at the link. Indeed,  1011.2642v1.pdf is a great paper. It uses Octonions in place of Quaternions for the construction of Ramanujan graphs and describes a wonderful breakthrough in creating small graphs with large girth. Peter Sarnak’s initial reaction to the new paper was: “wow”.

Here is a link to a paper entitled “Octonions” by John Baez, that appeared in Bull. AMS.

### Some background:

Let $G$ be a $k$-regular graph with girth $g$ where $g$ is an odd integer. Continue reading

# Benoît’s Fractals

Mandelbrot set

Benoît Mandelbrot passed away a few dayes ago on October 14, 2010. Since 1987, Mandelbrot was a member of the Yale’s mathematics department. This chapterette from my book “Gina says: Adventures in the Blogosphere String War”   about fractals is brought here on this sad occasion.

A little demonstration of Mandelbrot’s impact: when you search in Google for an image for “Mandelbrot” do not get pictures of Mandelbrot himself but rather pictures of Mandelbrot’s creation. You get full pages of beautiful pictures of Mandelbrot sets

.

Benoit Mandelbrot (1924-2010)

Modeling physics by continuous smooth mathematical objects have led to the most remarkable achievements of science in the last centuries. The interplay between smooth geometry and stochastic processes is also a very powerfull and fruitful idea. Mandelbrot’s realization of the prominence of fractals and his works on their study can be added to this short list of major paradigms in mathematical modeling of real world phenomena.

## Fractals

Fractals are beautiful mathematical objects whose study goes back to the late 19th century. The Sierpiński triangle and the Koch snowflake are early examples of fractals which are constructed by simple recursive rules. Continue reading

# Itamar Pitowsky: Probability in Physics, Where does it Come From?

I came across a videotaped lecture by Itamar Pitowsky given at PITP some years ago on the question of probability in physics that we discussed in two earlier posts on randomness in nature (I, II). There are links below to the presentation slides, and to  a video of the lecture.

A little over a week ago on Thursday, Itamar,  Oron Shagrir, and I sat at our little CS cafeteria and discussed this very same issue.  What does probability mean? Does it just represent human uncertainty? Is it just an emerging mathematical concept which is convenient for modeling? Do matters change when we move from classical to quantum mechanics? When we move to quantum physics the notion of probability itself changes for sure, but is there a change in the interpretation of what probability is?  A few people passed by and listened, and it felt like this was a direct continuation of conversations we had while we (Itamar and I; Oron is much younger) were students in the early 70s. This was our last meeting and Itamar’s deep voice and good smile are still with me.

In spite of his illness of many years Itamar looked in good shape. A day later, on Friday, he met with a graduate student working on connections between philosophy and computer science.  Yet another exciting new frontier. Last Wednesday Itamar passed away from sudden complications related to his illness.

Itamar was a great guy; he was great in science and great in the humanities, and he had an immense human wisdom and a modest, level-headed way of expressing it. I will greatly miss him.

Here is a link to a Condolence page for Itamar Pitowsky

 Probability in physics: where does it come from?

## Itamar Pitowsky

### Dept. of Philosophy, The Hebrew University of Jerusalem

The application of probability theory to physics began in the 19th century with Maxwell’s and Boltzmann’s explanation of the properties of gases in terms of the motion of their constituent molecules. Now the term probability is not a part of the (classical) theory of particle motion; so what does it mean, and where does it come from? Boltzmann thought to reduce the meaning of probability in physics to that of relative frequency. Thus, eg., we never find a container of gas in normal circumstances (equilibrium) with all of its molecules on the right hand side. Now, suppose we could prove this from the principles of mechanics- that a dynamical system with a huge number of particles almost never gets into a state with all its particles on one side. Then, to say that such an event has a vanishing probability would simply mean (and not only imply) that it is very rare.I shall explain Boltzmann’s program and assumptions in some detail, and why, in spite of its intuitive appeal, it ultimately fails. We shall also discuss why quantum mechanics with its “built in” concept of probability does not help much, and review some alternatives, as time permits.

Additional resources for this talk: video.

(Here is the original link to the PIPS lecture) My post entitled Amazing possibilities  about various fundamental limitations stated by many great minds that turned out to be wrong, was largely based on examples provided by Itamar.

# Randomness in Nature II

In a previous post we presented a MO question by Liza about randomness:

What is the explanation of the apparent randomness of high-level phenomena in nature?

1. Is it accepted that these phenomena are not really random, meaning that given enough information one could predict it? If so isn’t that the case for all random phenomena?

2. If there is true randomness and the outcome cannot be predicted – what is the origin of that randomness? (is it a result of the randomness in the micro world – quantum phenomena etc…)

Before I give the floor to the commentators, I would like to mention a conference on this topic that took place in Jerusalem a year ago. The title was “The Probable and the Improbable: The Meaning and Role of Probability in Physics” and the conference was in honor of Itamar Pitowsky. Let me also mention that  the Wikipedia article on randomness is also a good resource.

Here are some of the answers offered here to Liza’s question.

One way to think about what it means to say that a physical process is “random” is to say that there is no algorithm which predicts its behavior precisely which runs significantly faster than the process itself. Morally I think this should be true of many “high-level” phenomena. Continue reading

# When Noise Accumulates

I wrote a short paper entitled “when noise accumulates” that contains the main conceptual points (described rather formally) of my work regarding noisy quantum computers.  Here is the paper. (Update: Here is a new version, Dec 2010.) The new exciting innovation in computer science conference in  Beijing seemed tailor made for this kind of work, but the paper did not make it that far. Let me quote the first few paragraphs. As always, remarks are welcome!

From the introduction: Quantum computers were offered by Feynman and others and formally described by Deutsch, who also suggested that they can outperform classical computers. The idea was that since computations in quantum physics require an exponential number of steps on digital computers, computers based on quantum physics may outperform classical computers. A spectacular support for this idea came with Shor’s theorem that asserts that factoring is in BQP (the complexity class described by quantum computers).

The feasibility of computationally superior quantum computers is one of the most fascinating and clear-cut scientific problems of our time. The main concern regarding quantum-computer feasibility is that quantum systems are inherently noisy. (This concern was put forward in the mid-90s by Landauer, Unruh, and others.)

The theory of quantum error correction and fault-tolerant quantum computation (FTQC) and, in particular, the threshold theorem which asserts that under certain conditions FTQC is possible, provides strong support for the possibility of building quantum computers.

However, as far as we know, quantum error correction and quantum fault tolerance (and the highly entangled quantum states that enable them) are not experienced in natural quantum processes. It is therefore not clear if computationally superior quantum computation is necessary to describe natural quantum processes.

We will try to address two closely related questions. The first is, what are the properties of quantum processes that do not exhibit quantum fault tolerance and how to formally model such processes. The second is, what kind of noise models cause quantum error correction and FTQC to fail.

A main point we would like to make is that it is possible that there is  a systematic relation between the noise and the intended state of a quantum computer. Such a systematic relation does not violate linearity of quantum mechanics, and it is expected to occur in processes that do not exhibit fault tolerance.

Let me give an example: suppose that we want to simulate on a noisy quantum computer a certain bosonic state. The standard view of noisy quantum computers asserts that under certain conditions this can be done up to some error that is described by the computational basis. In contrast, the type of noise we expect amounts to having a mixed state between the intended bosonic state and other bosonic states (that represent the noise).

Criticism: A criticism expressed by several readers of an early version of this paper is that no attempt is made to motivate the conjectures from a physical point of view and that the suggestions seem “unphysical.” What can justify the assumption that a given error lasts for a constant fraction of the entire length of the process? If a noisy quantum computer at a highly entangled state has correlated noise between faraway qubits as  we suggest, wouldn’t it allow signaling faster than the speed of light?

I had sort of a mixed reaction toward this criticism. On the one hand I think that it is important and may be fruitful  to examine various models of noise while putting the physics aside. Nevertheless, I added a  brief discussion of  some physical aspects. Here it is: