Gaussian Process training with input noise

In standard Gaussian Process regression input locations are assumed to be noise free. We present a simple yet effective GP model for training on input points corrupted by i.i.d. Gaussian noise. To make computations tractable we use a local linear expansion about each input point. This allows the input noise to be recast as output noise proportional to the squared gradient of the GP posterior mean. The input noise variances are inferred from the data as extra hyperparameters. They are trained alongside other hyperparameters by the usual method of maximisation of the marginal likelihood. Training uses an iterative scheme, which alternates between optimising the hyperparameters and calculating the posterior gradient. Analytic predictive moments can then be found for Gaussian distributed test points. We compare our model to others over a range of different regression problems and show that it improves over current methods.

Toward automated generation of parametric BIMs based on hybrid video and laser scanning data

Only very few constructed facilities today have a complete record of as-built information. Despite the growing use of Building Information Modelling and the improvement in as-built records, several more years will be required before guidelines that require as-built data modelling will be implemented for the majority of constructed facilities, and this will still not address the stock of existing buildings. A technical solution for scanning buildings and compiling Building Information Models is needed. However, this is a multidisciplinary problem, requiring expertise in scanning, computer vision and videogrammetry, machine learning, and parametric object modelling. This paper outlines the technical approach proposed by a consortium of researchers that has gathered to tackle the ambitious goal of automating as-built modelling as far as possible. The top level framework of the proposed solution is presented, and each process, input and output is explained, along with the steps needed to validate them. Preliminary experiments on the earlier stages (i.e. processes) of the framework proposed are conducted and results are shown; the work toward implementation of the remainder is ongoing.

Monolithically integrated semiconductor optical amplifier and electroabsorption modulator with dual-waveguide spot-size converter input and output

We have demonstrated an electroabsorption modulator and semiconductor optical amplifier monolithically integrated with novel dual-waveguide spot-size converters (SSC) at the input and output ports for low-loss coupling to a planar light-guide circuit silica waveguide or cleaved single-mode optical fibre. The device was fabricated by means of selective-area MOVPE growth, quantum well intermixing and asymmetric twin waveguide technologies with only a three-step low-pressure MOVPE growth. For the device structure, in the SOA/EAM section, a double ridge structure was employed to reduce the EAM capacitances and enable high bit-rate operation. In the SSC sections, buried ridge structure (BRS) was incorporated. Such a combination of ridge, ATG and BRS structure is reported for the first time in which it can take advantage of easy processing of the ridge structure and the excellent mode characteristic of BRS. At the wavelength range of 1550-1600 nm, lossless operation with extinction ratios of 25 dB dc and more than 10 GHz 3 dB bandwidth is successfully achieved, The beam divergence angles of the input and output ports of the device are as small as 8.0 degrees x 12.6 degrees, resulting in 3.0 dB coupling loss with a cleaved single-mode optical fibre.

Monolithically integrated laser diode and electroabsorption modulator with dual-waveguide spot-size converter input and output

We have demonstrated a 1.60 mu m ridge-structure laser diode and electroabsorption modulator monolithically integrated with buried-ridge-structure dual-waveguide spot-size converters at the input and output ports for low-loss coupling to a cleaved single-mode optical fibre by means of selective area growth and asymmetric twin waveguide technologies. The devices emit in single transverse and quasi-single longitudinal modes with a side mode suppression ratio of 25.6 dB. These devices exhibit 3 dB modulation bandwidth of 15.0 GHz and modulator extinction ratios of 14.0 dB dc. The output beam divergence angles of the spot-size converter in the horizontal and vertical directions are as small as 7.3 degrees x 10.6 degrees, respectively, resulting in 3.0 dB coupling loss with a cleaved single-mode optical fibre.

Portfolio of Compositions investigating ensemble improvisation, distributed and real-time scoring, and multi-nodal input and output

This output is a collection of compositions which explore issues of ensemble improvisation, ensemble management and orchestration, real-time and distributed scoring, multi-nodal inputs and outputs, and animated and graphic notation. Compositions include: Activities I; tutti, duet, trio, solo, quartet; Lewitt Notations I; Webwork I; and Sometimes I feel the space between people (voices) in terms of tempos. These compositions are presented in computer animated scores which are synchronized through the network and subject to real-time modification and control. They can be performed by ensembles distributed over large physical spaces connected by the network. The scores for these compositions include software which displays the animations to the performers, software to structure and disseminate score events, and triggering software that allows the control of a performance to be distributed. Scores can also include live electronics which are coordinated with graphic events.

Efficiency assessment of electricity power generation on Ireland’s Single Electricity Market by Data Envelopment Analysis

The efficiency of generation plants is an important measure for evaluating the operating performance. The objective of this paper is to evaluate electricity power generation by conducting an All-Island-Generator-Efficiency-Study (AIGES) for the Republic of Ireland and Northern Ireland by utilising a Data Envelopment Analysis (DEA) approach. An operational performance efficiency index is defined and pursued for the year 2008. The economic activities of electricity generation units/plants examined in this paper are characterized by numerous input and output indicators. Constant returns to scale (CRS) and variable returns to scale (VRS) type DEA models are employed in the analysis. Also a slacks based analysis indicates the level of inefficiency for each variable examined. The findings from this study provide a general ranking and evaluation but also facilitate various interesting efficiency comparisons between generators by fuel type.

On Variable Selection of a Nonlinear Non-parametric System with a Limited Data Set: A Stepwise Algorithm

his paper considers a problem of identification for a high dimensional nonlinear non-parametric system when only a limited data set is available. The algorithms are proposed for this purpose which exploit the relationship between the input variables and the output and further the inter-dependence of input variables so that the importance of the input variables can be established. A key to these algorithms is the non-parametric two stage input selection algorithm.

Avaliação de desempenho dos municípios portugueses pelo método data envelopment analysis

Dissertação de mest., Economia Regional e Desenvolvimento Local, Faculdade de Economia, Univ. do Algarve, 2011

O uso do data envelopment analysis para avaliação e melhoria de desempenho no setor bancário: estudo de caso

O presente estudo aborda o uso do Data Envelopment Analysis (DEA) para avaliar e melhorar o desempenho no setor bancário. Consiste num estudo de caso aplicado a um dos maiores bancos portugueses. O seu objetivo, é avaliar a eficiência relativa das 333 agências bancárias que formam um dos dois departamentos comerciais do banco. Pretende-se identificar boas práticas e verificar a sua aplicabilidade nas unidades menos eficientes por forma a contribuir para a melhoria do desempenho global da instituição. Ao mesmo tempo, comparam-se os resultados de eficiência obtidos com o desempenho das unidades de negócio no cumprimento dos seus objetivos comerciais. Procura-se, assim, analisar a eventual existência de correlação entre eficiência e eficácia, ou seja, se as agências bancárias mais eficientes são também as mais eficazes. Para o efeito, é construído um modelo DEA que, considera em simultâneo, vários inputs e outputs. No modelo, as variáveis de input agregam os custos das unidades de negócio e as variáveis de output, os seus principais proveitos e alguns dos aspetos estratégicos que o banco pretende maximizar. Os resultados obtidos, identificam não só as agências bancárias menos eficientes, como assinalam aquelas que sendo semelhantes e eficientes, lhes podem servir de referência para a melhoria do seu desempenho. Tendo em conta, o caráter formativo que se pretende para o estudo, e de modo a facilitar a aceitação pelos decisores, procura-se verificar a exequibilidade das propostas apresentadas. O documento conclui pela existência, nas unidades de negócio analisadas, de uma correlação positiva fraca entre eficiência e eficácia, ainda assim estatisticamente significante. Por último, assinala a importância da metodologia DEA, enquanto medida de complementaridade de outras técnicas de controlo de gestão, em particular nas organizações que adotam a gestão por objetivos.