Roberta Sinatra

From Santa Fe Institute Events Wiki

About Roberta

Me one year ago

Hi! I'm a second year PhD student in Physics, at ATP group of University of Catania (Italy), my advisor being Prof. Vito Latora. Since I started my MSc in Physics, I have always been fascinated by statistical mechanics, nonlinear dynamics and complex systems. In particular, connections with other disciplines like biology, economy and social sciences have captured my attention. For my MSc thesis I worked on statistical methods to convert sequences of symbols into networks, and I applied them to proteins. Afterwards, I have started focusing my interests on the study of structure, functions and evolution of complex networks, from both theoretical and empirical points of view. Before starting my PhD, I have spent some months at BIFI, University of Zaragoza (Spain), where I worked on Evolutionary Game Theory, studying the evolution of fairness by means of an the Ultimatum Game played on graphs, and at the School of Mathematical Sciences of Queen Mary College (London, UK), where I focused on entropy rate in Markov chains and on coupled "man-made" networks. Currently, for my PhD thesis, I am working on diffusion processes on graphs, in particular on biased random walks and on the influence of graph structure on dynamical processes. Part of this project is developed in collaboration with Prof. H.J. Jensen of the Institute for Mathematical Sciences of the Imperial College (London), and with Prof. J. Gomez-Gardenes of the Universidad Rey Juan Carlos (Madrid).

I dedicate some of my time also to a "side" project concerning neuroscience: I study the topological properties of connectivity patterns derived by EEG measurements in humans performing different tasks. I work on this last project in collaboration with Fondazione Santa Lucia, Roma.

In my spare time, I like traveling, reading, "discovering" old movies, cooking and making gastronomic experiments, playing volleyball, learning foreign languages, ...


  • E-mail: roberta dot sinatra at ct dot infn dot it


What are your main interests? Feel free to include a "pie in the sky" big idea!

I'm interested in understanding how topology affects dynamical processes and how one can design networks or diffusion and dynamical processes in order to get the best performance from the system. For example, which is the best process to explore a network just with local information? Or which is the best way to spread information? Which topology favours evolution and cooperation when agents move on a network and interact? Are there stationary states for all usual processes taking place on networks? Although these questions may sound very theoretical and merely speculative, they have a broad range of applications and turn out to be very useful in many systems and problems. I try to face and answer these questions not only with models and with numerical simulations, but also and moreover with analytical approaches.

We can discuss about all this next June!

What sort of expertise can you bring to the group?

I have experience with numerical methods and complex network analysis. Since I have a background as a theoretical physicist, I am also able to deal with proofs and mathematical tools. In particular, I use the approach typical of statistical mechanics to face a broad range of problems. This modus operandi necessitates of some knowledges of statistics, probability, stochastic processes.

I can also program in C, C++, Matlab and I have some skills with Pajek (when I don't lose my patience with it…).

What do you hope to get out of the CSSS?

I think that the best way to face a research (and not only! ;) ) problem, to get inspiration and new ideas, is to interact and have nice discussions with other people, whether they work on the same research field or not. And what is the best place but this summer school to do it? ;) Needless to say, I expect also to have a good time with all other people taking part in the school. Looking forward to meeting all of you!

Do you have any possible projects in mind for the CSSS?

If you didn't get bored to read all I wrote till now, you probably understood I am "obsessed" with network :P During the summer school, I would like to develop an evolutionary model on graph I have already started to work on a couple of months ago. In order to give some solid bases to the model, discussions and interactions with people with a background in social sciences and in biology would be very useful and motivating. More specifically, the model deals on how genomes of walkers exploring a network evolve and change according to the spatial structure they move on. Affinity or dissimilarity between genomes is determined by the structure of the graph, and their analysis allows to detect ecological clustering or social communities.

As technical issue, I will dedicate part of my time during the school to learn python, since I have read many participants are able with it and could give me useful tips and how-to.