Friday, May 30, 2008

The program is over - Friday 30th

Yesterday went on with a nice dinner at the restaurant Esperanza on Birger Jarlsgatan.

Here is a photo of the proceedings (courtesy of Petter Holme):

From left to right, clockwise: Meesoon Ha, Petter Holme, Mikko Alava, Haijun Zhou, Matti Peltomäki, Sumedha Sumedha, and Sang-hoon Lee. Thanks to Laila of Nordita for the choice....

Then we did some science to celebrate the end. A discussion session about the physics of peer-to-peer neworks was held in the morning of the Friday.

Supriya Krishnamurthy presented first the issue from a physicst's viewpoint, of networks and dynamics on dynamic networks. Then György Dan gave a computer science response, of what is important, and why, in the study of self-* (wild card) processes. A lively discussion ensued. Can a theoretical physicist show entrepreneurial spirit and contribute (while not becoming rich)?

A PS: their presentations are now on the program web page.

We thank Nordita in general for making the program possible, and hospitality, the participants (also of the workshop) for the scientific atmosphere, and the local Nordita folks for assisting, helping and many other little things that made PhysDIS possible. Erik Aurell has been ill for the last week: to a quick recovery!

Thursday, May 29, 2008

Thursday talk, 29th of May

Sumedha Sumedha talked today about her recent research on biophysics: how to understand transcription factor binding and its relation to how cells operate. The basic question boils down to an interplay of geometry, chemical potential, and the polymer physics (entropy of loops) of say DNA. For simple model systems the outcome is a funny model of random graphs with a link/degree restriction per vertex. Which, can be studied analytically and out comes a percolation transition which can be first or second order.

Later today, we will have a dinner...

3rd last blog post, Thursday morning

The SMDIS program is crawling closer and closer to its eventual end.

The previous Tuesday was without any formal program thanks to some parallel bioinformatics related activity here in Stockholm. Inspired by the nice spring/summer weather this gave way to a working day. Yesterday, Meesoon Ha presented some of her recent work on one-dimensional extensions of the Asymmetric Exclusion Process (ASEP). The main issue at hand is that the ASEP is one of the standard paradigms of non-equilibrium systems, and being a representation of the Kardar-Parisi-Zhang problem (or the Burgers' equation) exhibits shocks and nonlinearities. One can then modify the ASEP to look at various transitions that correspond most often to shock-formation in the KPZ, or to seek for new physics - like if one in the ASEP lets the particles to jump not by one lattice site but to the next cluster of particles.

Monday, May 26, 2008

Monday Markets (the 26th of May)

The life of the SMDIS program returned after an eventful weekend (with a boat-trip to the archipelago of Stockholm and Erik Aurell getting slightly ill) to normal: work and a talk.

Today, Matteo Marsili gave an overview of recent work on multi-asset markets, empirical data, and theoretical modelling. In practice this means that if a lot of investors behave in the same way, that is use a similar strategy to optimize their portfolios, then this starts to have a feed-back effect with the way the market itself acts. As often in these models (Matteo presented two approaches, one "physics-style" and another "economy-style". It was not quite clear how to get from the more complicated latter to the former...) one has the possibility of a transition between a state with large fluctuations and a state with small fluctuations.

Tomorrow we will have an only-working day thanks to a simultaneous bioinformatics event here in Stockholm.

Friday, May 23, 2008

Friday 23th and Google

Today, we heard Sang-Hoon Lee talk about his research on networks. The main idea is to use Google to find out about effective social networks, using the joint hits (e.g. for Mikko Alava and Albert Einstein, they are found as exact keywords together 7 times). So a graph is established with weights, and then one can study its properties and development and compare with other possible measures for the same ensemble connected by the "Google-network". Results on the U.S. senators were presented....

Report on Thursday 22nd

The main event of the Thursday was the talk of Satya Majumdar on the relation between the Integer Partitioning Problem and the statistics of extremes. By the right choice for the energy of the state ("number") one can map the problem to the statistics of a Bose gas. Then, even an interpretation can be found in how the Gumbel distribution a long ago found in mathematics arises. Many extensions could be foreseen (many of which were already resolved by Satya).

Meanwhile, the weather got better and warmer.

Thursday, May 22, 2008

A Wednesday passed (the 21st)

Yesterday we had one talk, by Matteo Palassini. He discussed some recent work in Barcelona on the p53 regulatory network: the appearance of noise-induced oscillations in its dynamics due to number (of molecules) fluctuations. A coupling of the eigenvalues of the dynamical system and the noise gives rise to this. Matteo had a bit of PowerPoint trouble, but at the end it was resolved by the traditional means of a (white)board which then enlivened the audience...

... meanwhile the weather got better.

Wednesday, May 21, 2008

Tuesday 20th passed...

We have had a perfectly ordinary day with just one talk by prof. Bing-Hong Wang, about three different models of co-operation and adaptation and their spatial behavior. He overviewed the Prisoner's Dilemma Game, the Snowdrift Game, and finally the Naming Game. The two first models exhibit various properties as regards the graph structure's and the pay-off matrix' influence on the likelihood of co-operation taking place. In the Naming Game it seems that asymmetric dynamics and emphasizing the role of "hubs" in networks both create interesting effects.

Tuesday, May 20, 2008

Back to (program) business - Monday 16th

After some rest and relaxation a normal Monday followed: work and a talk by prof. Chin-Kun Hu on protein folding. Here, one of the essential issues is to understand the folding pathways. The talk presented us some results that illustrated the challenges how to model the problem and how to design experiments by force microscopy, such that they work by giving the right sequence. It appears that here simple statistical physics have a lot to give, but also some distance to go.

Saturday, May 17, 2008

Saturday, last workshop day - May 17th

Last night, social activities interfered with science as we had a very nice workshop dinner at Restaurant Godthem in Djurgården.

Today instead, the last day was celebrated (during an otherwise rainy day) with a lot of program. The audience reached again 40-45 at maximum (the early morning being an exception). The first part was on computer science/physics -related issues, with three long contributions. First Johan Håstad told us why (eg.) MAX-SAT is so very difficult to approximate (actually you can as well try a random assignment...). Then, Ashish Sabharwal and Alexander Hartmann educated the audience more on why Constraint Satisfaction Problems (CSP) are interesting. Ashish talked about how to sample the number of solutions cleverly, and Alex about analyzing clustering in COL and 3-SAT and its relation to what Local Search does. The structure of the CSP solution (and energy) landscape was also addressed in short talks by Frederico Ricci-Tersenghi, Florent Krzakala, and Lenka Zdeborova.
They all presented very recent results.

Finally, the Saturday was finished with some mixed topic -talks, and with Matteo Marsili discussing How to Be Lucky, or how to park (in Marseille).

Hopefully all participants liked this event, in particular those who came to Stockholm/Nordita only for the workshop and not also for the program.

Friday, May 16, 2008

Long days, long talks at SMDIS workshop - Friday 16th

Though the program is crowded there is still a sizeable crowd following the workshop, at 16 + \epsilon amounting to about 50 people.

Today's program has more long, 40 minute talks than yesterday (Thursday). We heard then from Riccardo Zecchina how to understand random Steiner trees using statistical mechanics (using a combination of belief propagation and the cavity equation approach) and then Jeremy Stribling discussed how to put together a distributed file system (or how to share data) "without problems": how to create a protocol for doing that in the P2P-spirit.

The Friday program has contained no less than five long talks. First Massimo Vergassola outlined the role of non-coding RNAs in bacteria. Then, Haijun Zhou overviewed his recent work on dynamics on networks: how the scale-free character influences opinion model dynamics (it is much dependent on the degree distribution exponent) and how such dynamics can be coupled with network evolution to create scale-free -looking networks. Scott Kirkpatrick told us how to treat the Sudoku puzzle as a challenging optimization problem (and how to solve it by various techniques). This was contrasted nicely by Ilkka Niemelä, whose presentation contained a lot of information on industrial SAT problems, how they are to be solved efficiently (and why), and what the real challenges in SAT are in the future. Finally, Neil Gershenfeld finished the day by a tale about how to program bits and atoms. Or, more precisely, how to inject programming into our surroundings, and how to learn from physics and biology to make such a holistic view work, including a mixture of nano-fabrication and really ubiquitous computing.

Thursday, May 15, 2008

Wednesday and Thursday of SMDIS

We have now entered the heavy part of the program. Wednesday saw an interesting talk in the morning by György Dan of the KTH here in Stockholm. He discussed dynamical analysis of the workings of various peer-to-peer systems. Then the day continued with the K-SAT related licentiate thesis defense of John Ardelius (SICS, Stockholm). This was aptly carried out via Skype since his opponent Bart Selman (Cornell) is not yet here.

And, on Thursday morning commenced the program. The many talks are on the web program, just to mention that Supriya Krishnamurthy was the first speaker. Tonight the main social event is a visit to the opera. We had about 50 people typically present.

Tuesday, May 13, 2008

Monday and Tuesday at Nordita

In view of the upcoming workshop, more and more people are gathering to the program. This includes eg. Meesoon Ha from KAIST, Matteo Palassini (Barcelona), Joachim Krug, and Chin-Kun Hu, Academia Sinica, Taipei, and so forth.

The early week has seen on Monday the Nobel Forum talk of Massimo Vergassola on Biological Strategies of Motility and further today some biophysics -oriented ones by Krug and Heiko Rieger (Dynamics of microbial evolution, and, Angiogenesis and vascular network modeling in a growing tumor, respectively). The program is of course a bit heavy with two daily talks, but maybe this is a good practice for the workshop?

Friday, May 9, 2008

Final day of the first week, Friday

The program continues to enjoy nice weather.

Then, we had a quite pedagogical and interesting talk by Satya Majumdar on the distribution of the largest eigenvalue of some classes of random matrices. The crux of the matter is that one can compute this exactly, and the result turns out to be exponential in the square of N, matrix size (sometimes linear in N, in the Wishart case) times a constant. The constant depends on the ensemble, and on where one looks at that on the real axis. The fact that seems to have caught the interest of some participants is the fact that the Coulomb gas mapping used to compute finally the distribution is the most clear and simple of all available ones. See recent papers by Satya, including Olav Tirkkonen pointed out some connections to communications theory.

Of course, the day was finished with wine and cheese (thanks Anne and Laila!) to celebrate the first week.

Thursday, May 8, 2008

Thursday, May 8th, and the SMDIS

Today, we had two interesting deviations from "normal program life".

First, prof. Joakim Enwall entertained the audience with a summary of the languages along the "silk roads", between the West and China. What were those, actually, and what happened to them? Of course this is inter-connected with the modern world and in particular with the fact that SMDIS has Chinese participants.

Then Luca Peliti continued, this time with a colloquium on the influenza virus and its epidemological modelling. It appears to be necessary to understand empirical data via the tendency of the virus to mutate during the course of an epidemic (which among others makes it possible for the same person to fall sick from apparently the same virus again and again...). Usual SIR-models can modified to account for the varying dynamic strain populations of the virus, but this is not yet adequate and Luca discussed extensively the reasons and future directions related to that.

Chinese SMDIS guests (Bing-Hong Wang and Haijun Zhou) and Heiko Rieger (Saarbrucken) have now joined us.

Wednesday, May 7, 2008

Wednesday 7th at SMDIS

In today's first talk Dr. Paolo Muratore-Ginanneschi (University of Helsinki) discussed the scaling properties of the two-dimensional Navier-Stokes equation, for which the existence and uniqueness of solution still remains without a proof. Paolo pointed out differences between two different theories of the problem, leading to the conclusion that the renormalization theory of two-dimensional fully-developed turbulence in the Navier-Stokes equation needs to be revisited - still work to be done by the participants and colleagues.

In the afternoon, Prof. Olav Tirkkonen (Helsinki University of Technology, Theory and NokiaCo. ) talked about his ongoing work on self-organizing communication networks. In particular, his idea is to draw lessons from statistical physics of synchronization phenomena and apply them to engineering problems related to ad hoc communication networks. In a very lively ex tempore discussion session following Olav's talk, several ideas and connections to earlier work rose from the audience that Tirkkonen and co-workers can exploit - even during the programme (hopefully).

(Thanks to Matti Peltomäki for a concise summary).

Tuesday, May 6, 2008

Tuesday 6th at SMDIS

Today, more people showed up, so that the non-Nordic visitors were Paolo Muratore-Ginanneschi, Olav Tirkkonen (well, from HUT), Luca Peliti, Satya Majumdar, Matteo Marsili, at the end of the day.

The main highlight was the seminar of Luca Peliti (copies of the slides should be available via the web site here and for the talks in general ) on Sequence evolution in phenotypic space: An overview. The main idea is to look at statistical mechanics models on transcription factor binding, and think about the space of possible simple models, and what that implies of how organisms work and how experiment should be interpreted. Related work has been done in particular by the group of Michael Lässig, recently.

Monday, May 5, 2008

Monday 5th at Nordita

Today the program started. And it did it so via an opening talk by Erik Aurell. He explained the background of SMDIS and its funny relation to the sister program at KITPC, Beijing. Together these account for three months of science that at the end may help to define what is this "SMDIS". Erik went through a number of issues discussed in China to explain to the non-experts some of the challenges that a physicist faces in particular in Constraint Satisfaction Problems (3-SAT is a keyword), like what to say about the UNSAT-phase. Then, the constraint density is so high that an instance of the (say) 3-SAT problem can not be solved, it has a "positive energy" in that the optimal assignment leaves some constraints violated. It is a challenge to find physics-based approaches to beat the computer science techniques for proving unsolvability.

Finally, Erik listed some (pet) directions that really deal with "SMDIS" or at least contain the DIS-part and look for statistical mechanics for new ideas - or did so already. Examples arise in distributed Peer-to-Peer systems, in overlay network management (dynamics), and in designing distributed algorithms for evolving, complex networks.

Tomorrow we shall have a first real talk (sorry Erik) by Luca Peliti (Naples), then on Wednesday Olav Tirkkonen (TKK, Helsinki).

Saturday, May 3, 2008

On the eve of the Nordita program

At this point, the upcoming weeks look interesting. There has been some last minute extra interest concerning both the program and the mid-term workshop on May 15-17. This event will be in particular packed with three full days of program... but, while one should give the participants a chance to work on science, we also will be kept busy with one or two daily talks for the bread-and-butter part of the schedule before and after the workshop.

It should also be acknowledged already, that we have had extra support for PhD students (from Sweden, the Access Foundation, from Finland, the National Graduate School for Materials Physics - which obviously also includes the physics of complex systems...).

The real scientific program will kick-off on Tuesday 6th by a talk from Luca Peliti from Naples. A glance at the rest of the week tells that in addition to "program events" there will also be talks by participants at other occasions and institutions in Stockholm: a welcome sign. Updates of the scientific life will be published here regularly.

Speaking of welcomes on Monday the 5th there will be a welcome event for the people who have reached Nordita already by then.