Pavlos Kazakopoulos

We propose an interpretation for the meets and joins in the lattice of experimental propositions of a physical theory, answering a question of Birkhoff and von Neumann in [1]. When the lattice is atomistic, it is isomorphic to the lattice of definably closed sets of a finitary relational structure in First Order Logic. In terms of mapping experimen...

Morocco has a rich and diverse cultural heritage including 35 ancient medina’s. Despite these important assets, Morocco is facing some serious stagnation in reaching their tourism ambitions as have been defined in their Vision 2020. To identify their relevance for (urban) tourism, it is crucial to understand and to include the historic and multi-la...

We present and test a sequential learning algorithm for the prediction of human mobility that leverages large datasets of sequences to improve prediction accuracy, in particular for users with a short and non-repetitive data history such as tourists in a foreign country. The algorithm compensates for the difficulty of predicting the next location w...

Construction of the experts. (PDF)

Sleeping Experts Exponential Weights forecaster. (PDF)

Experts fragmentation. Average length of a fragment in an expert sequence. Almost 95% of the experts have an average fragment length of 5 hours or less. The high degree of sequence fragmentation, due to inherent irregularities in the sequence of connection events in the CDR data, distorts the statistics of the user’s Markov model relative to the re...

Eta grid and optimal Eta. (A) Average prediction accuracy as a function of η for a logarithmic grid of 30 points equally log-spaced between 10−2 and 103. The average prediction accuracy reaches a peak at η ≈ 3 and drops only very slightly after that, even for very high learning rates. This is due to the abundance of experts in the ensemble. Severel...

Entropy. Entropy per element estimate using the Lempel-Ziv estimator for a random sample of the test sequences. In an overwhelming majority of cases, the estimate has not stabilised before the sequence ends. This indicates that the mobility patterns that the sequence represents are not regular, or the sequence is too short to detect regularities. (...

Transience of test sequences. (PDF)

Grid-based urbanization velocity analysis of remote sensing imagery is used to measure urban growth rates. However, it remains unclear how critical the spatial resolution of the imagery is to such grid-based approaches. This research therefore investigated how urbanization velocity estimates respond to different spatial resolutions, as determined b...

In the area of Big Data an data science organizations intends to initiate a series of experiments to probe the field, to assess technologies and to develop skills. There are many diverse objectives associated to these experiments, such as: • Surveying and assessing how to open up the current datasets (data access and availability); • Increasing the...

We present and test a sequential learning algorithm for the short-term prediction of human mobility. This novel approach pairs the Exponential Weights forecaster with a very large ensemble of experts. The experts are individual sequence prediction algorithms constructed from the mobility traces of 10 million roaming mobile phone users in a European...

In the advent of a pervasive presence of location sharing services researchers gained an unprecedented access to the direct records of human activity in space and time. This paper analyses geo-located Twitter messages in order to uncover global patterns of human mobility. Based on a dataset of almost a billion tweets recorded in 2012 we estimate vo...

Linear receivers offer a low complexity option for multi-antenna communication systems. Therefore, understanding the outage behavior of the corresponding SINR is important in a fading mobile environment. In this paper we introduce a large deviations method, valid nominally for a large number M of antennas, which provides the probability density of...

The explosion of demand for ultra-high information transmission rates over the last decade has necessitated the usage of increasingly high light intensities for fiber optical transmissions. As a result, the fiber non-linearities need to be treated non-perturbatively. Similar analyses in the past have focused on the effects of non-linearities on exi...

Linear receivers offer a low complexity option for multi-antenna communication systems. Therefore, understanding the outage behavior of the corresponding SINR is important in a fading mobile environment. In this paper we introduce a large deviations method, valid nominally for a large number M of antennas, which provides the probability density of...

Using a large deviations approach we calculate the probability distribution of the mutual information of MIMO channels in the limit of large antenna numbers. In contrast to previous methods that only focused at the distribution close to its mean (thus obtaining an asymptotically Gaussian distribution), we calculate the full distribution, including...

Using a large deviations approach we calculate the probability distribution of the mutual information of MIMO channels in the limit of large antenna numbers. In contrast to previous methods that only focused to the distribution close to its most probable value, thus obtaining an asymptotically Gaussian distribution, we calculate the full distributi...

A large deviations approach is introduced, which calculates the probability density and outage probability of the MIMO mutual information, and is valid for large antenna numbers N. In contrast to previous asymptotic methods that only focused on the distribution close to its most probable value, this methodology obtains the full distribution, includ...

We calculate the Lyapunov exponent for the non-Hermitian Zakharov-Shabat eigenvalue problem corresponding to the attractive nonlinear Schrödinger equation with a Gaussian random pulse as an initial value function. Using an extension of the Thouless formula to non-Hermitian random operators, we calculate the corresponding average density of states....

We calculate the Lyapunov exponent for the non-Hermitian Zakharov-Shabat eigenvalue problem corresponding to the attractive non-linear Schroedinger equation with a Gaussian random pulse as initial value function. Using an extension of the Thouless formula to non-Hermitian random operators, we calculate the corresponding average density of states. W...

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Physics, 2005. Includes bibliographical references (p. 167-173). In this thesis we study four-dimensional supersymmetric gauge theories and their string theory realization. After an introduction and a brief review of some basic concepts at the interface of gauge theories and string th...

We construct all connected toric phases of the recently discovered Yp,q quivers and show their IR equivalence using Seiberg duality. We also compute the R and global U(1) charges for a generic toric phase of Yp,q.

We construct a new infinite family of = 1 quiver gauge theories which can be Higgsed to the Yp,q quiver gauge theories. The dual geometries are toric Calabi-Yau cones for which we give the toric data. We also discuss the action of Seiberg duality on these quivers, and explore the different Seiberg dual theories. We describe the relationship of thes...

We study four dimensional N=1 gauge theories that arise on the worldvolume of D3-branes probing complex cones over del Pezzo surfaces. Global symmetries of the gauge theories are made explicit by using a correspondence between bifundamental fields in the quivers and divisors in the underlying geometry. These global symmetries are hidden, being unbr...

We compute the NSVZ beta functions for N = 1 four-dimensional quiver theories arising from D-brane probes on singularities, complete with anomalous dimensions, for a large set of phases in the corresponding duality tree. While these beta functions are zero for D-brane probes, they are non-zero in the presence of fractional branes. As a result there...