An assumption-driven case-specific model editor

A case sensitive intelligent model editor has been developed for constructing consistent lumped dynamic process models and for simplifying them using modelling assumptions. The approach is based on a systematic assumption-driven modelling procedure and on the syntax and semantics of process,models and the simplifying assumptions.

Use of the q-Gaussian mutation in evolutionary algorithms

This paper proposes the use of the q-Gaussian mutation with self-adaptation of the shape of the mutation distribution in evolutionary algorithms. The shape of the q-Gaussian mutation distribution is controlled by a real parameter q. In the proposed method, the real parameter q of the q-Gaussian mutation is encoded in the chromosome of individuals and hence is allowed to evolve during the evolutionary process. In order to test the new mutation operator, evolution strategy and evolutionary programming algorithms with self-adapted q-Gaussian mutation generated from anisotropic and isotropic distributions are presented. The theoretical analysis of the q-Gaussian mutation is also provided. In the experimental study, the q-Gaussian mutation is compared to Gaussian and Cauchy mutations in the optimization of a set of test functions. Experimental results show the efficiency of the proposed method of self-adapting the mutation distribution in evolutionary algorithms.

Expectation Propagation with Factorizing Distributions: A Gaussian Approximation and Performance Results for Simple Models

We discuss the expectation propagation (EP) algorithm for approximate Bayesian inference using a factorizing posterior approximation. For neural network models, we use a central limit theorem argument to make EP tractable when the number of parameters is large. For two types of models, we show that EP can achieve optimal generalization performance when data are drawn from a simple distribution.

On the validity of the dispersion model of hepatic drug elimination when intravascular transit time densities are long-tailed

The dispersion model with mixed boundary conditions uses a single parameter, the dispersion number, to describe the hepatic elimination of xenobiotics and endogenous substances. An implicit a priori assumption of the model is that the transit time density of intravascular indicators is approximated by an inverse Gaussian distribution. This approximation is limited in that the model poorly describes the tail part of the hepatic outflow curves of vascular indicators. A sum of two inverse Gaussian functions is proposed as ail alternative, more flexible empirical model for transit time densities of vascular references. This model suggests that a more accurate description of the tail portion of vascular reference curves yields an elimination rate constant (or intrinsic clearance) which is 40% less than predicted by the dispersion model with mixed boundary conditions. The results emphasize the need to accurately describe outflow curves in using them as a basis for determining pharmacokinetic parameters using hepatic elimination models. (C) 1997 Society for Mathematical Biology.

Quantum and classical fidelities for Gaussian states

We examine the physical significance of fidelity as a measure of similarity for Gaussian states by drawing a comparison with its classical counterpart. We find that the relationship between these classical and quantum fidelities is not straightforward, and in general does not seem to provide insight into the physical significance of quantum fidelity. To avoid this ambiguity we propose that the efficacy of quantum information protocols be characterized by determining their transfer function and then calculating the fidelity achievable for a hypothetical pure reference input state. (c) 2007 Optical Society of America.

Mean anisotropy of homogeneous Gaussian random fields and anisotropic norms of linear translation-invariant operators on multidimensional integer lattices

Sensitivity of output of a linear operator to its input can be quantified in various ways. In Control Theory, the input is usually interpreted as disturbance and the output is to be minimized in some sense. In stochastic worst-case design settings, the disturbance is considered random with imprecisely known probability distribution. The prior set of probability measures can be chosen so as to quantify how far the disturbance deviates from the white-noise hypothesis of Linear Quadratic Gaussian control. Such deviation can be measured by the minimal Kullback-Leibler informational divergence from the Gaussian distributions with zero mean and scalar covariance matrices. The resulting anisotropy functional is defined for finite power random vectors. Originally, anisotropy was introduced for directionally generic random vectors as the relative entropy of the normalized vector with respect to the uniform distribution on the unit sphere. The associated a-anisotropic norm of a matrix is then its maximum root mean square or average energy gain with respect to finite power or directionally generic inputs whose anisotropy is bounded above by a≥0. We give a systematic comparison of the anisotropy functionals and the associated norms. These are considered for unboundedly growing fragments of homogeneous Gaussian random fields on multidimensional integer lattice to yield mean anisotropy. Correspondingly, the anisotropic norms of finite matrices are extended to bounded linear translation invariant operators over such fields.

Adapted Gaussian basis sets for atoms from Li through Xe generated with the generator coordinate Hartree-Fock method

Utiliza-se o método coordenada geradora Hartree-Fock para gerar bases Gaussianas adaptadas para os átomos de Li (Z=3) até Xe (Z=54). Neste método, integram-se as equações de Griffin-Hill-Wheeler-Hartree-Fock através da técnica de discretização integral. Comparam-se as funções de ondas geradas neste trabalho com as funções de ondas Roothaan-Hartree-Fock de Clementi e Roetti (1974) e com outros conjuntos de bases relatados na literatura. Para os átomos estudados aqui, os erros em nossas energias totais relativos aos limites numéricos Hartree-Fock são sempre menores que 7,426 milihartree.

Tomographic imaging with a scanning laser ophtalmoscop

Retinal imaging with a confocal scaning laser Ophthalmoscope (cSLO) involves scanning a small laser beam over the retina and constructing an image from the reflected light. By applying the confocal principle, tomographic images can be produced by measuring a sequence of slices at different depths. However, the thickness of such slices, when compared with the retinal thickness, is too large to give useful 3D retinal images, if no processing is done. In this work, a prototype cSLO was modified in terms hardware and software to give the ability of doing the tomographic measurements with the maximum theoretical axial resolution possible. A model eye was built to test the performance of the system. A novel algorithm has been developed which fits a double Gaussian curve to the axial intensity profiles generated from a stack of images slices. The underlying assumption is that the laser light has mainly been reflected by two structures in the retina, the internal limiting membrane and the retinal pigment epithelium. From the fitted curve topographic images and novel thickness images of the retina can be generated. Deconvolution algorithms have also been developed to improve the axial resolution of the system, using a theoretically predicted cSLO point spread function. The technique was evaluated using measurements made on a model eye, four normal eyes and seven eyes containing retinal pathology. The reproducibility, accuracy and physiological measurements obtained, were compared with available published data, and showed good agreement. The difference in the measurements when using a double rather than a single Gaussian model was also analysed.

Machine learning Vasicek model calibration with Gaussian processes

In this article, we calibrate the Vasicek interest rate model under the risk neutral measure by learning the model parameters using Gaussian processes for machine learning regression. The calibration is done by maximizing the likelihood of zero coupon bond log prices, using mean and covariance functions computed analytically, as well as likelihood derivatives with respect to the parameters. The maximization method used is the conjugate gradients. The only prices needed for calibration are zero coupon bond prices and the parameters are directly obtained in the arbitrage free risk neutral measure.

Digital-domain self-calibration technique for video-rate pipeline A/D converters using Gaussian white noise

Electronics Letters Vol.38, nº 19

Low-voltage low-power CMOS analogue circuits for Gaussian and uniform noise generation

IEEE International Symposium on Circuits and Systems, MAY 25-28, 2003, Bangkok, Thailand. (ISI Web of Science)

Efficient digital self-calibration of video-rate pipeline ADCs using white gaussian noise

Proceedings of IEEE, ISCAS 2003, Vol.I, pp. 877-880

Improving Assumption based Distributed Belief Revision

Belief revision is a critical issue in real world DAI applications. A Multi-Agent System not only has to cope with the intrinsic incompleteness and the constant change of the available knowledge (as in the case of its stand alone counterparts), but also has to deal with possible conflicts between the agents’ perspectives. Each semi-autonomous agent, designed as a combination of a problem solver – assumption based truth maintenance system (ATMS), was enriched with improved capabilities: a distributed context management facility allowing the user to dynamically focus on the more pertinent contexts, and a distributed belief revision algorithm with two levels of consistency. This work contributions include: (i) a concise representation of the shared external facts; (ii) a simple and innovative methodology to achieve distributed context management; and (iii) a reduced inter-agent data exchange format. The different levels of consistency adopted were based on the relevance of the data under consideration: higher relevance data (detected inconsistencies) was granted global consistency while less relevant data (system facts) was assigned local consistency. These abilities are fully supported by the ATMS standard functionalities.

Brownian bridge and other path-dependent Gaussian processes vectorial simulation

The iterative simulation of the Brownian bridge is well known. In this article, we present a vectorial simulation alternative based on Gaussian processes for machine learning regression that is suitable for interpreted programming languages implementations. We extend the vectorial simulation of path-dependent trajectories to other Gaussian processes, namely, sequences of Brownian bridges, geometric Brownian motion, fractional Brownian motion, and Ornstein-Ulenbeck mean reversion process.

Machine learning Gaussian short rate

Dissertação para obtenção do Grau de Doutor em Estatística e Gestão do Risco