Proposals SURF@IFISC 2019

1.  Big data and language variation 

Advisor: David Sánchez 

Using Twitter corpora, it has been shown that significant geographical variations of word usage appear worldwide. The aim of this SURF project is to investigate whether those variations can be analyzed with the Zipf’s law and whether the Zipf’s law can present deviations between different dialect areas of English. We will consider two major English varieties (British and American) and will analyze whether the diatopic polarization changes with time and user's connectivity. This project will be developed in collaboration with linguists. 

2. Complexity in quantum spin chains 

Advisor: Gian Luca Giorgi

Exactly solvable quantum spin chains represent a powerful tool to study many-body problems in the quantum regime. Given a network of quantum units (the spin chain) it is possible to build different classical networks with link weights given by two-body quantum correlations, such as entanglement, discord, mutual information. In this project, making use of the analytical solution available for Isinglike quantum models, we propose to study the complexity properties of such classical networks in order to infer the critical features and the underlying physics of the quantum spin chain itself. 

3. Power grid stability under demand stress and cascading failures 

Advisors: Pere Colet and Damià Gomila

The power grid is, arguably, the largest socio-technical system in the world. Stable operation requires the synchronization of the power plants and a precise balance between generation and consumption. The balance is not easy to achieve due to the random character of (part of) the load and the increasing use of renewable sources which are subject to uncontrollable factors, such as wind or sunlight. Large demand changes, such as those occurring at touristic areas in high season further stress the system. In this project we will study the synchronization and stability of a prototypical power grid when subject to large demand fluctuations, the probability of eventual blackouts and cascading failures on the network. 

4. Chemical fronts in fluids 

Advisor: Emilio Hernández-García

Many chemical and biological reactions occur in fluids, so that there is a complex interplay between the chemical transformation dynamics and the motion, often turbulent, of the fluid. For some types of reactions, relevant for example to flame propagation or to plankton growth in the sea, sharp fronts (e.g. interfaces separating the reacted from the unreacted fluid) form, and propagate in a way that can be understood from novel methodologies, based on dynamical systems and chaos theory, known under the name of ‘burning manifolds theory’. The aim of this project is to explore several examples of front propagation in simplified fluid flows, some of them of relevance in marine biology processes.

5. Topological transport in hybrid superconducting systems 

Advisor: Llorenç Serra

Topological transport systems are characterized by the  presence of current-carrying states along the edges and boundaries  of the material. The project will address the theoretical description  of topological transport in a class of materials, quantum-anomalous Hall insulators with induced superconductivity [1]. In particular, the  conductance of 2D strips having a normal-superconductingnormal junction in different topological phases will be calculated using the  complex-k techniques of Ref. [2] and compared with recent experiments [1]. The main objective of this SURF project is to become familiar with the phenomenology and theoretical models. As potential extensions we may consider the following: i) study arbitrarily oriented magnetizations of the material, ii) including orbital effects of the perpendicular component of the magnetic field, iii) study thermal effects when the contacts are injecting quasiparticles of different energy distributions. 

[1] He, Q. L. et al.; Science 2017, 357, 294–299. 

[2] Osca, J.; Serra, L.; Phys. Rev. B 2018, 98, 121407.

6. Machine Learning in Quantum Transport for topological nanostructures 

Advisors: Rosa López and David Sánchez

A machine learning approach to solve electronic quantum transport equations of quasi-one-dimensional nanostructures is applied for the case Majorana physics is present. The transmission coefficients of nanowire systems were computed to provide training and test data sets to the learning process by considering different nanowire sizes, geometries and interaction strengths. This technique allows us to a fast prediction for new linear conductance matrices (for heat and charge currents) for a large data set different devices. Eventually our method could serve as a method to identify the relevance of interactions in the Majorana mode formation in quasi-one-dimensional nanowires. 

7. Big Data, memes, and information diffusion in online social networks 

Advisors: Jose Ramasco and Sandro Meloni

Memes are topical units spreading in online social networks. They are formed by myriads of messages, replies, and forwards, containing from simple information diffusion events to opinions and views on the topic discussed. The entrance of memes in the network usually occurs simultaneously from different points (by different users) and spread afterward in a local way in the network. Memes fight over users' attention mediated by what the network visibility allows. The objective here is to model this type of diffusion and compare it with empirical data.

This web uses cookies for data collection with a statistical purpose. If you continue browsing, it means acceptance of the installation of the same.

More info I agree