Category Archives: Computer Science and Optimization

Lawler-Kozdron-Richards-Stroock’s combined Proof for the Matrix-Tree theorem and Wilson’s Theorem

wilson  curvature

David Wilson and a cover of Shlomo’s recent book “Curvature in mathematics and physics”

A few weeks ago, in David Kazhdan’s basic notion seminar, Shlomo Sternberg gave a lovely presentation Kirchho ff and Wilson via Kozdron and Stroock. The lecture is based on the work presented in the very recent paper by Michael J. Kozdron,  Larissa M. Richards, and Daniel W. Stroock: Determinants, their applications to Markov processes, and a random walk proof of Kirchhoff’s matrix tree theorem. Preprint, 2013. Available online at arXiv:1306.2059.

Here is the abstract:

Kirchhoff’s formula for the number of spanning trees in a connected graph  is over 150 years old. For example, it says that if c_2, \dots, c_n are the nonzero  eigenvalues of the Laplacian, then the number k of spanning trees is k= (1/n)c_2\cdots c_n. There are many proofs.  An algorithm due to Wilson via loop erased random walks produces such a tree, and Wilson’s theorem is that all spanning trees are produced by his algorithm with equal probability. Hence,  after the fact, we know that Wilson’s algorithm produces any given tree with probability 1/k.  A proof due to Lawler, using the Green’s function, shows directly that Wilson’s algorithm has the probability 1/k  of producing any given spanning tree, thus simultaneously proving Wilson’s theorem and Kirchhoff’s formula. Lawler’s proof has been considerably simplified by Kozdron and Stroock. I plan to explain their proof. The lecture will be completely self-contained, using only Cramer’s rule from linear algebra.

(Here are also lecture notes of the lecture by Ron Rosenthal.)

Here is some background.

The matrix-tree theorem

The matrix tree theorem asserts that the number of rooted spanning trees of a connected graph G  is the product of the non-zero eigenvalues of L(G), the Laplacian of G.

Suppose that G has n vertices. The Laplacian of G is the matrix whose (i,i)-entry is the degree of the ith vertex, and its (i,j) entry for i \ne j is 0 if the ith vertex is not adjacent to the jth vertex, and -1 if they are adjacent. So  L(G)=D-A(G) where A(G) is the adjacency matrix of G, and D is a diagonal matrix whose entries are the degrees of the vertices.  An equivalent formulation of the matrix-tree theorem is that the number of spanning trees is the determinant of a matrix obtained from the Laplacian by deleting the j th row and j th column.

We considered a high dimensional generalization of the matrix tree theorem in these posts (I, II, III, IV).

How to generate a random spanning tree for a graph G?

Using the matrix-tree theorem

Method A: Start with an edge e \in G, use the matrix-tree theorem to compute the probability p_e that e belongs to a random spanning tree of G, take e with probability p_e. If e is taken consider the contraction G/e and if G is not taken consider the deletion G \backslash e and continue.

This is an efficient method to generate a random spanning tree according to the uniform probability distribution. You can extend it by assigning each edge a weight and chosing a tree with probability proportional to the product of its weights.

Random weights and greedy

Method B: Assign each edge a random real number between 0 and 1 and chose the spanning tree which minimizes the sum of weights via the greedy algorithm.

This is a wonderful method but it leads to a different probability distribution on random spanning trees which is very interesting!

The Aldous-Broder random walk method

Method C: The Aldous-Broder theorem. Start a simple random walk from a vertex of the graph until reaching all vertices, and take each edge that did not form a cycle with earlier edges. (Or, in other words, take every edge that reduced the number of connected components of the graph on the whole vertex set and visited edges.)

Amazingly, this leads to a random uniform spanning tree. The next method is also very amazing and important for many applications.

David Wilson’s algorithm

Method D: Wilson’s algorithm. Fix a vertex as a root. (Later the root will be a whole set of vertices, and a tree on them.) Start from an arbitrary vertex u not in the root and take a simple random walk until you reach the root. Next, erase all edges in cycles of the path created by the random walk so you will left with a simple path from  u to the root. Add this path to the root and continue!

Here is a link to Wilson’s paper! Here is a nice presentation by Chatterji  and Gulwani.

The Kadison-Singer Conjecture has beed Proved by Adam Marcus, Dan Spielman, and Nikhil Srivastava

…while we keep discussing why mathematics is possible…

The news

Adam Marcus, Dan Spielman, and Nikhil Srivastava posted a paper entitled “Interlacing Families II: Mixed Characteristic Polynomials and the Kadison-Singer Problem,” where they prove the 1959 Kadison-Singer conjecture.

(We discussed part I of “interlacing families” in this post about new Ramanujan graphs.  Looks like a nice series.)

The Kadison-Singer Conjecture

The Kadison-Singer conjecture refers to a positive answer to a question posed by Kadison and Singer: “They asked ‘whether or not each pure state of \cal B is the extension of some pure state of some maximal abelian algebra’ (where \cal B is the collection of bounded linear transformations on a Hilbert space.”) I heard about this question in a different formulation known as the “Bourgain-Tzafriri conjecture” (I will state it below) and the paper addresses a related well known discrepancy formulation by Weaver. (See also Weaver’s comment on the appropriate “quantum” formulation of the conjecture.)

Updates: A very nice post on the relation of the Kadison-Singer Conjecture  to foundation of quantum mechanics is in this post in  Bryan Roberts‘ blog Soul Physics. Here is a very nice post on the mathematics of the conjecture with ten interesting comments on the proof by Orr Shalit, and another nice post on Yemon Choi’s blog and how could I miss the very nice post on James Lee’s blog.. Nov 4, 2013: A new post with essentially the whole proof appeared on Terry tao’s blog, Real stable polynomials and the Kadison Singer Problem.

Update: A very nice blog post on the new result was written by  Nikhil Srivastava on “Windows on theory.” It emphasizes the discrapancy-theoretic nature of the new result, and explains the application for partitioning graphs into expanders.

The Bourgain-Tzafriri theorem and conjecture

Let me tell again (see this post about Lior, Michael, and Aryeh where I told it first) about a theorem of Bourgain and Tzafriri related to the Kadison-Singer conjecture.

Jean Bourgain and Lior Tzafriri considered the following scenario: Let a > 0 be a real number. Let A be a n \times n matrix with norm 1 and with zeroes on the diagonal. An s by s principal minor M is “good” if the norm of M is less than a.

Consider the following hypergraph F:

The vertices correspond to indices {}[n]=\{1,2,\dots,n\}. A set S \subset [n] belongs to F if the S \times S sub-matrix of M is good.

Bourgain and Tzafriri showed that for every a > 0 there is C(a) > 0 so that for every matrix A we can find S \in F so that |S| \ge C(a)n.

Moreover, they showed that for every nonnegative weights w_1,w_2,\dots w_n there is S \in F so that the sum of the weights in S is at least C(a) times the total weight. In other words, (by LP duality,) the vertices of the hypergraph can be fractionally covered by C(a) edges.

The “big question” is if there a real number C'(a) > 0 so that for every matrix M, {}[n] can be covered by C'(a) good sets. Or, in other words, if the vertices of F can be covered by C'(a) edges. This question is known to be equivalent to an old conjecture by Kadison and Singer (it is also known as the “paving conjecture”). In view of what was already proved by Bourgain and Tzafriri what is needed is to show that the covering number is bounded from above by a function of the fractional covering number. So if you wish, the Kadison-Singer conjecture had become a statement about bounded integrality gap. Before proving the full result, Marcus, Spielman and Srivastava gave a new proof of the Bourgain-Tzafriti theorem.

Additional references:

KADISON-SINGER MEETS BOURGAIN-TZAFRIRI by PETER G. CASAZZA, ROMAN VERSHYNIN,  The Kadison-Singer Problem in Mathematics and Engineering: A Detailed Account pdf, and many other recent publications by Pete Casazza.

Why is mathematics possible?

Spectacular advances in number theory

Last weeks we heard about two spectacular results in number theory.  As announced in Nature, Yitang Zhang proved that there are infinitely many pairs of consecutive primes (p_n, p_{n+1}) which are at most 70 million apart! This is a sensational achievement. Pushing 70 million to 2 will settle the ancient conjecture on twin primes, but this is already an extremely amazing breakthrough.  An earlier breakthrough came in 2005 when Daniel Goldston, János Pintz, and Cem Yıldırım proved that the gaps between consecutive primes p_{n+1}-p_n is infinitely often smaller than \sqrt {\log p_n} \log \log ^2 p_n.

Update: A description of Zhang’s work and a link to the paper can be found on Emmanuel Kowalski’s bloog Further update: A description of Zhang’s work and related questions and results can be found now in Terry Tao’s blog. Terry Tao also proposed a new polymath project aimed to reading Zhang’s paper and attempting to improve the bounds.

Harald Helfgott proved that every integer is the sum of three primes.  Here the story starts with Vinogradov who proved it for sufficiently large integers, but pushing down what “sufficiently large” is, and pushing up the computerized methods needed to take care of “small” integers required much work and ingenuity.

Why is Mathematics possible?

The recent news, and a little exchange of views I had with Boaz Barak, bring us back to the question: “Why is mathematics possible?” This is an old question that David Kazhdan outlined in a lovely 1999 essay “Reflection on the development of mathematics in the twentieth century.” The point (from modern view) is this: We know that mathematical statements can, in general, be undecidable.  We also know that a proof for a short mathematical statement can be extremely long. And we also know that even if a mathematical statement admits a short proof, finding such a proof can be computationally intractable. Given all that, what are the reasons that mathematics is at all possible?

It is popular to associate “human creativity” with an answer. The problem with incorrect (or, at least, incomplete) answers is that they often represent missed opportunities for better answers. I think that for the question “why is mathematics possible” there are opportunities (even using computational complexity thinking) to offer better answers.

Please offer your answers.

qstart poster


Physics, Computer Science, Mathematics, and Foundations’
views on quantum information

Inauguration conference for the Quantum Information Science Center (QISC),
Hebrew university of Jerusalem

Update: The news of our conference have made it to a big-league blog.

Update (July 2013): QStart was a very nice event- there were many interesting talks, and the speakers made the effort to have lectures accessible to the wide audience while discussing the cutting edge and at times technical matters.Streaming video of the talks is now available.

My Quantum Debate with Aram III

This is the third and last post giving a timeline and some non technical highlights from my debate with Aram Harrow.  

Where were we

After Aram Harrow and I got in touch in June 2011, and decided to have a blog debate towards the end of 2011, the first post in our debate describing my point of view was launched on January, 2012 and was followed by three posts by Aram. The discussion was intensive and interesting.  Here is a link to my 2011 paper that initiated the debate and to a recent post-debate presentation at MIT.

 Happy_Passover  Happy passover, readers!

Back to the debate: Conjecture C is shot down!


In addition to his three posts, Aram and Steve Flammia wrote a paper refuting one of my Conjectures (Conjecture C).  We decided to devote a post to this conjecture.

Quantum refutations and reproofs

Post 5, May 12, 2012. One of Gil Kalai’s conjectures refuted but refurbished

Niels Henrik Abel was the patron saint this time

The first version of the post started with this heartbreaking eulogy for Conjecture C. At the end most of it was cut away. But the part about Aram’s grandchildren was left in the post.

Eulogy for Conjecture C

(Gil; old version:) When Aram wrote to me, inn June 2011, and expressed willingness to publicly discuss my paper, my first reaction was to decline and propose having just private discussions. Even without knowing Aram’s superb track record in debates, I knew that I put my beloved conjectures on the line. Some of them, perhaps even all of them, will not last. Later, last December, I changed my mind and Aram and I started planning our debate. My conjectures and I were fully aware of the risks. And it was Conjecture C that did not make it.

A few words about Conjecture C

Conjecture C, while rooted in quantum computers skepticism, was a uniter and not a divider! It expressed our united aim to find a dividing line between the pre- and post- universal quantum computer eras.

Aram’s grandchildren and the world before quantum computers

When Aram’s grandchildren will ask him: “
Grandpa, how was the world before quantum computers?” he could have replied: “I hardly remember, but thanks to Gil we have some conjectures recording the old days, and then he will state to the grandchildren Conjectures 1-4 and the clear dividing line in terms of Conjecture C, and the grandchildren will burst in laughter about the old days of difficult entanglements.” Continue reading

Mittag-Leffler Institute and Yale, Winter 2005; Test your intuition: Who Played the Piano?

This is a little “flashback” intermission in my posts about my debate with Aram Harrow. This time I try to refer to Cris Moore’s question regarding  the motivation for my study. For the readers it gives an opportunity to win a $50 prize! 

Let me also bring to your attention an interesting discussion (starting here) between Peter Shor and me regarding smoothed Lindblad evolutions.

(Cris Moore, the debate’s very first comment!) I am also a little confused by Gil’s motivation for his conjectures.  (My response:)  To the best of my memory, my main motivation for skeptically studying quantum fault-tolerance was that I thought that this is a direction worth pursuing and that I had a shot at it.

micheldevoretposter (1)

While listening with Ravi Kannan to this 2002 lecture by Michel Devoret at Yale, I wondered if there are enough scientists working on the “mirage” side.

Flashback: Mittag-Leffler 2005

I started systematically thinking about quantum fault-tolerance in February 2005. There were several things that triggered my interest to the question in the previous fall and I decided to spend some time learning and thinking about it in our winter break.  One of those triggers was something Dorit Aharonov told me a few months earlier: she said that once, when she was telling her students about quantum computers, she suddenly had a feeling that maybe it was all just nonsense. Another trigger came from a former student who told me about a Polish scientist (whose name he could not remember) who wrote an article about impossibility of quantum error-correction. I thought that the lack of a quantum analog of the repetition code, and the unique properties of the majority function  in terms of sensitivity to noise that I studied with Itai Benjamini and Oded Schramm earlier could be a good starting point for looking skeptically at quantum computers.  

In our 2005 winter break, I spent two weeks at Yale and then additional two weeks at the Mittag-Leffler institute near Stockholm.  At Yale, I only had little time to think about quantum computers. I had to finish a survey article with Muli Safra about threshold phenomena (To a volume that Cris Moore and Allon Perkus were among the editors).  One of the last days in Yale we went to dinner with two guests, Chris Skinner who gave the colloquium talk, and Andrei Okounkov who visited me and gave a talk about partition enumeration and mirror symmetry. At the dinner Andrew Casson asked, out of the blue, if we think that quantum computers can be built and it almost seemed as if that Andrew was reading my mind on what I plan to work on the weeks to come. My answer there was the same as my answer now, that I tend to find it implausible.

Mittag-Leffler Institute February 2005 with Xavier Viennot and Alain Lascoux

In Sweden I spent most of my time on quantum fault-tolerance. I was jet-lagged and being jet-lagged in the Mittag-Leffler institute already worked for me once, when finding my subexponential randomized variant of the simplex algorithm was a substitute for sleeping some night in fall 1991 . In 2005 it was not as bad, I just came to my office very early in the morning and started working. And very early in the morning somebody was already playing the piano.

And who was playing the piano at the institute in the cold Swedish mornings of February 2005? The first reader to guess correctly, and convince me in a comment that she or he knows the answer without revealing it to everybody else will get $50. Continue reading

My Quantum Debate with Aram II

This is the second of three posts giving few of the non-technical highlights of my debate with Aram Harrow. (part I)

After Aram Harrow and I got in touch in June 2011, and decided to have a blog debate about quantum fault-tolerance towards the end of 2011, the first post in our debate was launched on January 30, 2012.  The first post mainly presented my point of view and it led to lovely intensive discussions. It was time for Aram’s reply and some people started to lose their patience.

(rrtucky) Is Aram, the other “debater”, writing a dissertation in Greek, as a reply?

Flying machines of the 21st century

Post II, February 6, 2011. First of three responses by Aram Harrow

Dave Bacon was the patron saint for Aram’s first post.

(Aram) There are many reasons why quantum computers may never be built…  The one thing I am confident of is that we are unlikely to find any obstacle in principle to building a quantum computer.

(Aram) If you want to prove that 3-SAT requires exponential time, then you need an argument that somehow doesn’t apply to 2-SAT or XOR-SAT. If you want to prove that the permanent requires super-polynomial circuits, you need an argument that doesn’t apply to the determinant. And if you want to disprove fault-tolerant quantum computing, you need an argument that doesn’t also refute fault-tolerant classical computing.

From the discussion

Why not yet? Boaz set a deadline


(Boaz Barak could [you] explain a bit about the reasons why people haven’t been able to build quantum computers with more than a handful of qubits yet? Continue reading