Regularized Online Mixture of Gaussians for Background Subtraction

Mixture of Gaussians (MoG) modelling [13] is a popular approach to background subtraction in video sequences. Although the algorithm shows good empirical performance, it lacks theoretical justification. In this paper, we give a justification for it from an online stochastic expectation maximization (EM) viewpoint and extend it to a general framework of regularized online classification EM for MoG with guaranteed convergence. By choosing a special regularization function, l1 norm, we derived a new set of updating equations for l1 regularized online MoG. It is shown empirically that l1 regularized online MoG converge faster than the original online MoG .

Improving parameter learning of Bayesian nets from incomplete data

This paper addresses the estimation of parameters of a Bayesian network from incomplete data. The task is usually tackled by running the Expectation-Maximization (EM) algorithm several times in order to obtain a high log-likelihood estimate. We argue that choosing the maximum log-likelihood estimate (as well as the maximum penalized log-likelihood and the maximum a posteriori estimate) has severe drawbacks, being affected both by overfitting and model uncertainty. Two ideas are discussed to overcome these issues: a maximum entropy approach and a Bayesian model averaging approach. Both ideas can be easily applied on top of EM, while the entropy idea can be also implemented in a more sophisticated way, through a dedicated non-linear solver. A vast set of experiments shows that these ideas produce significantly better estimates and inferences than the traditional and widely used maximum (penalized) log-likelihood and maximum a posteriori estimates. In particular, if EM is adopted as optimization engine, the model averaging approach is the best performing one; its performance is matched by the entropy approach when implemented using the non-linear solver. The results suggest that the applicability of these ideas is immediate (they are easy to implement and to integrate in currently available inference engines) and that they constitute a better way to learn Bayesian network parameters.

Site choices in recreational demand: a matter of utility maximization or regret minimization?

This paper compares the Random Regret Minimization and the Random Utility Maximization models for determining recreational choice. The Random Regret approach is based on the idea that, when choosing, individuals aim to minimize their regret – regret being defined as what one experiences when a non-chosen alternative in a choice set performs better than a chosen one in relation to one or more attributes. The Random Regret paradigm, recently developed in transport economics, presents a tractable, regret-based alternative to the dominant choice paradigm based on Random Utility. Using data from a travel cost study exploring factors that influence kayakers’ site-choice decisions in the Republic of Ireland, we estimate both the traditional Random Utility multinomial logit model (RU-MNL) and the Random Regret multinomial logit model (RR-MNL) to gain more insights into site choice decisions. We further explore whether choices are driven by a utility maximization or a regret minimization paradigm by running a binary logit model to examine the likelihood of the two decision choice paradigms using site visits and respondents characteristics as explanatory variables. In addition to being one of the first studies to apply the RR-MNL to an environmental good, this paper also represents the first application of the RR-MNL to compute the Logsum to test and strengthen conclusions on welfare impacts of potential alternative policy scenarios.

Stated choices and benefit estimates in the context of traffic calming schemes:Utility maximization, regret minimization, or both?

This paper proposes a discrete mixture model which assigns individuals, up to a probability, to either a class of random utility (RU) maximizers or a class of random regret (RR) minimizers, on the basis of their sequence of observed choices. Our proposed model advances the state of the art of RU-RR mixture models by (i) adding and simultaneously estimating a membership model which predicts the probability of belonging to a RU or RR class; (ii) adding a layer of random taste heterogeneity within each behavioural class; and (iii) deriving a welfare measure associated with the RU-RR mixture model and consistent with referendum-voting, which is the adequate mechanism of provision for such local public goods. The context of our empirical application is a stated choice experiment concerning traffic calming schemes. We find that the random parameter RU-RR mixture model not only outperforms its fixed coefficient counterpart in terms of fit-as expected-but also in terms of plausibility of membership determinants of behavioural class. In line with psychological theories of regret, we find that, compared to respondents who are familiar with the choice context (i.e. the traffic calming scheme), unfamiliar respondents are more likely to be regret minimizers than utility maximizers. © 2014 Elsevier Ltd.

Characterisations of the weak expectation property

We use representations of operator systems as quotients to deduce various characterisations of the weak expectation property (WEP) for C∗ -algebras. By Kirchberg’s work on WEP, these results give new formulations of Connes’ embedding problem.

Dual-Diversity Combining for Constrained Resource Allocation and Throughput Maximization in OFDMA Networks

Al Rawi, Anas F., Emiliano Garcia-Palacios, Sonia Aissa, Charalampos C. Tsimenidis, and Bayan S. Sharif. "Dual-Diversity Combining for Constrained Resource Allocation and Throughput Maximization in OFDMA Networks." In Vehicular Technology Conference (VTC Spring), 2013 IEEE 77th, pp. 1-5. IEEE, 2013.

Rate Maximization of Decode-and-Forward Relaying Systems with RF Energy Harvesting

We consider a three-node decode-and-forward (DF) half-duplex relaying system, where the source first harvests RF energy from the relay, and then uses this energy to transmit information to the destination via the relay. We assume that the information transfer and wireless power transfer phases alternate over time in the same frequency band, and their time fraction (TF) may change or be fixed from one transmission epoch (fading state) to the next. For this system, we maximize the achievable average data rate. Thereby, we propose two schemes: (1) jointly optimal power and TF allocation, and (2) optimal power allocation with fixed TF. Due to the small amounts of harvested power at the source, the two schemes achieve similar information rates, but yield significant performance gains compared to a benchmark system with fixed power and fixed TF allocation.

Joint Information and Jamming Beamforming for Secrecy Rate Maximization in Cognitive Radio Networks

In this paper, we consider the secure beamforming design for an underlay cognitive radio multiple-input singleoutput broadcast channel in the presence of multiple passive eavesdroppers. Our goal is to design a jamming noise (JN) transmit strategy to maximize the secrecy rate of the secondary system. By utilizing the zero-forcing method to eliminate the interference caused by JN to the secondary user, we study the joint optimization of the information and JN beamforming for secrecy rate maximization of the secondary system while satisfying all the interference power constraints at the primary users, as well as the per-antenna power constraint at the secondary transmitter. For an optimal beamforming design, the original problem is a nonconvex program, which can be reformulated as a convex program by applying the rank relaxation method. To this end, we prove that the rank relaxation is tight and propose a barrier interior-point method to solve the resulting saddle point problem based on a duality result. To find the global optimal solution, we transform the considered problem into an unconstrained optimization problem. We then employ Broyden-Fletcher-Goldfarb-Shanno (BFGS) method to solve the resulting unconstrained problem which helps reduce the complexity significantly, compared to conventional methods. Simulation results show the fast convergence of the proposed algorithm and substantial performance improvements over existing approaches.

Polarisation analysis of VUV synchrotron radiation emitted from a bending magnet source in the energy range 20-50 eV: A comparison between measurements and theoretical predictions

The results of a study to characterise the polarisation properties of the photon beam emerging from beamline 5D, mounted on a bending magnet source at the Synchrotron Radiation Source, Daresbury Laboratory, are presented. The expectation values for the Stokes parameters corresponding to the light transmitted by the beamline have been calculated by combining ray-tracing and optical methods. The polarisation of the light at the source is modified both by the beamline geometry and by the reflections at the optical components. Although it is often assumed that the polarising properties of grazing incidence optics are negligible, this assumption leads to rather inaccurate results in the VUV region. A study of the reflectivity shows that even at incidence angles (theta(i) = 80-85degrees) which are far from the Brewster angle (theta(B) similar to 45degrees for VUV and soft X-ray radiation) the residual changes in the amplitudes of the reflected light can result in non-negligible polarisation effects. Furthermore, reflection at grazing incidence gives rise to a substantial change in the phase, and this has the effect of rotating the elliptically polarised state. Theoretical Stokes parameters have been compared with full polarisation measurements obtained using a reflection polarimeter in the energy range 20-40 eV. (C) 2003 Elsevier B.V. All rights reserved.

Classical computation with quantum systems

As semiconductor electronic devices scale to the nanometer range and quantum structures (molecules, fullerenes, quantum dots, nanotubes) are investigated for use in information processing and storage, it, becomes useful to explore the limits imposed by quantum mechanics on classical computing. To formulate the problem of a quantum mechanical description of classical computing, electronic device and logic gates are described as quantum sub-systems with inputs treated as boundary conditions, outputs expressed.is operator expectation values, and transfer characteristics and logic operations expressed through the sub-system Hamiltonian. with constraints appropriate to the boundary conditions. This approach, naturally, leads to a description of the subsystem.,, in terms of density matrices. Application of the maximum entropy principle subject to the boundary conditions (inputs) allows for the determination of the density matrix (logic operation), and for calculation of expectation values of operators over a finite region (outputs). The method allows for in analysis of the static properties of quantum sub-systems.