Multi-level visualisation using Gaussian process latent variable models

Projection of a high-dimensional dataset onto a two-dimensional space is a useful tool to visualise structures and relationships in the dataset. However, a single two-dimensional visualisation may not display all the intrinsic structure. Therefore, hierarchical/multi-level visualisation methods have been used to extract more detailed understanding of the data. Here we propose a multi-level Gaussian process latent variable model (MLGPLVM). MLGPLVM works by segmenting data (with e.g. K-means, Gaussian mixture model or interactive clustering) in the visualisation space and then fitting a visualisation model to each subset. To measure the quality of multi-level visualisation (with respect to parent and child models), metrics such as trustworthiness, continuity, mean relative rank errors, visualisation distance distortion and the negative log-likelihood per point are used. We evaluate the MLGPLVM approach on the ‘Oil Flow’ dataset and a dataset of protein electrostatic potentials for the ‘Major Histocompatibility Complex (MHC) class I’ of humans. In both cases, visual observation and the quantitative quality measures have shown better visualisation at lower levels.

1-Tb/s DWDM amplification in a fiber optical parametric amplifier

We report less than 1-dB cross-talk penalty for 26 DWDM channels modulated at 43.7 Gb/s RZ-DPSK when amplified by a fiber optical parametric amplifier showing compatibility with high-capacity (> 1 Tb/s) communication systems. © 2010 Optical Society of America.

1-Tb/s DWDM long-haul transmission employing a fiber optical parametric amplifier

In this letter, we report the performance of a fiber optical parametric amplifier (OPA) when used as a source or intermediate node amplifier in a dense wavelength-division-multiplexed (DWDM) long-haul transmission testbed with 26 DWDM channels modulated at 43.7-Gb/s return-to-zero differential phase-shift keying. In both scenarios, we demonstrate similar performance to an erbium-doped fiber amplifier. This shows the OPAs compatibility with high-capacity (>1 Tb/s) long-haul communication systems.

Digital fiber nonlinearity compensation:toward 1-Tb/s transport

The world is connected by a core network of long-haul optical communication systems that link countries and continents, enabling long-distance phone calls, data-center communications, and the Internet. The demands on information rates have been constantly driven up by applications such as online gaming, high-definition video, and cloud computing. All over the world, end-user connection speeds are being increased by replacing conventional digital subscriber line (DSL) and asymmetric DSL (ADSL) with fiber to the home. Clearly, the capacity of the core network must also increase proportionally. © 1991-2012 IEEE.

Semisupervised learning of hierarchical latent trait models for data visualization

Recently, we have developed the hierarchical Generative Topographic Mapping (HGTM), an interactive method for visualization of large high-dimensional real-valued data sets. In this paper, we propose a more general visualization system by extending HGTM in three ways, which allows the user to visualize a wider range of data sets and better support the model development process. 1) We integrate HGTM with noise models from the exponential family of distributions. The basic building block is the Latent Trait Model (LTM). This enables us to visualize data of inherently discrete nature, e.g., collections of documents, in a hierarchical manner. 2) We give the user a choice of initializing the child plots of the current plot in either interactive, or automatic mode. In the interactive mode, the user selects "regions of interest," whereas in the automatic mode, an unsupervised minimum message length (MML)-inspired construction of a mixture of LTMs is employed. The unsupervised construction is particularly useful when high-level plots are covered with dense clusters of highly overlapping data projections, making it difficult to use the interactive mode. Such a situation often arises when visualizing large data sets. 3) We derive general formulas for magnification factors in latent trait models. Magnification factors are a useful tool to improve our understanding of the visualization plots, since they can highlight the boundaries between data clusters. We illustrate our approach on a toy example and evaluate it on three more complex real data sets. © 2005 IEEE.

73.7 Tb/s (96X3x256-Gb/s) mode-division-multiplexed DP-16QAM transmission with inline MM-EDFA

We show transmission of a 73.7 Tb/s (96x3x256-Gb/s) DP-16QAM modedivision- multiplexed signal over 119km of few-mode fiber with inline multi-mode EDFA, using 6x6 MIMO digital signal processing. The total demonstrated net capacity is 57.6 Tb/s (SE 12 bits/s/Hz). © 2012 OSA.

Economic efficiency of mobile latent heat storages

In a pilot project an optimized mobile latent heat storage based on a system available on the market has been tested at Fraunhofer Institute for Environmental, Safety and Energy Technology. Initially trials were conducted with the aim of optimizing the process of charging and discharging. A specifically constructed test rig at the incineration trials centre at the institute allowed charging and discharging procedures of the mobile latent heat storage with adjustable parameters. In addition an evaluation model was constructed to further optimize the heat exchanger systems. In conclusion the prototype of the mobile latent heat storage was tested in practical operation. The economic and technical feasibility of heat transportation was shown if not utilized waste heat is available. © 2014 The Authors.

73.7 Tb/s (96 x 3 x 256-Gb/s) mode-divisionmultiplexed DP-16QAM transmission with inline MM-EDFA

Transmission of a 73.7 Tb/s (96x3x256-Gb/s) DP-16QAM mode-division- multiplexed signal over 119km of few-mode fiber transmission line incorporating an inline multi mode EDFA and a phase plate based mode (de-)multiplexer is demonstrated. Data-aided 6x6 MIMO digital signal processing was used to demodulate the signal. The total demonstrated net capacity, taking into account 20% of FEC-overhead and 7.5% additional overhead (Ethernet and training sequences), is 57.6 Tb/s, corresponding to a spectral efficiency of 12 bits/s/Hz. © 2012 Optical Society of America.

Many Facet, Common Latent Trait Polytomous IRT Model and EM Algorithm

2000 Mathematics Subject Classification: 91E45.

4 Tb/s transmission reach enhancement using 10 × 400 Gb/s super-channels and polarization insensitive dual band optical phase conjugation

In this paper, we experimentally demonstrate the benefit of polarization insensitive dual-band optical phase conjugation for up to ten 400 Gb/s optical super-channels using a Raman amplified transmission link with a realistic span length of 75 km. We demonstrate that the resultant increase in transmission distance may be predicted analytically if the detrimental impacts of power asymmetry and polarization mode dispersion are taken into account.

Semi-supervised Gaussian process latent variable model with pairwise constraints

In machine learning, Gaussian process latent variable model (GP-LVM) has been extensively applied in the field of unsupervised dimensionality reduction. When some supervised information, e.g., pairwise constraints or labels of the data, is available, the traditional GP-LVM cannot directly utilize such supervised information to improve the performance of dimensionality reduction. In this case, it is necessary to modify the traditional GP-LVM to make it capable of handing the supervised or semi-supervised learning tasks. For this purpose, we propose a new semi-supervised GP-LVM framework under the pairwise constraints. Through transferring the pairwise constraints in the observed space to the latent space, the constrained priori information on the latent variables can be obtained. Under this constrained priori, the latent variables are optimized by the maximum a posteriori (MAP) algorithm. The effectiveness of the proposed algorithm is demonstrated with experiments on a variety of data sets. © 2010 Elsevier B.V.

A kockázat mint látens fogalom (Risk - A latent concept)

A kockázat statisztikai értelemben közvetlenül nem mérhető, azaz látens fogalom éppen úgy, mint a gazdasági fejlettség, a szervezettség vagy az intelligencia. Mi bennünk a közös? A kockázat is komplex fogalom, több mérhető tényezőt foglal magában, és bár sok tényezőjét mérjük, fel sem tételezzük, hogy pontos eredményt kapunk. Ebben a megközelítésben az elemző kezdettől fogva tudja, hogy hiányos az ismerete. Ezt Bélyácz [2011[ nyomán úgy is megfogalmazhatjuk: „A statisztikusok tudják, hogy valamit éppen nem tudnak.” / === / From statistical point of view risk, like economic development is a latent concept. Typically there is no one number which can explicitly estimate or project risk. Variance is used as a proxy in finance to measure risk. Other professions are using other concepts for risk. Underwriting is the most important step in insurance business to analyse exposure. Actuaries evaluate average claim size and the probability of claim to calculate risk. Bayesian credibility can be used to calculate insurance premium combining frequencies and empirical knowledge, as a prior. Different types of risks can be classified into a risk matrix to separate insurable risk. Only this category can be analysed by multivariate statistical methods, which are based on statistical data. Sample size and frequency of events are relevant not only in insurance, but in pension and investment decisions as well.

The relationship between spirituality, knowledge and tuberculosis (TB) medication adherence among African Americans and Haitians

Tuberculosis (TB) is an infectious disease and nonadherence to medication can lead to new cases, multi-drug resistant TB, or potential death. Additionally, healthcare professionals and individuals with TB’s knowledge of the disease and medication adherence are crucial for successful completion of medication therapy. Patient education is one of the most important aspects of care provided in healthcare settings (CDC, 1994). TB tends to disproportionately affect minority and economically disadvantaged patient populations. The purpose of this mixed method study was to explore the relationship between spirituality, knowledge, and TB medication adherence among African Americans and Haitians. The primary research question was: What is the relationship between spirituality, knowledge and TB medication adherence among African Americans and Haitians? Quantitative data were gathered from 33 questionnaires and analyzed by two ANOVAs and four chi square analyses. The null hypothesis was not rejected; there was not a statistically significant relationship between spirituality and TB medication adherence (p =.208) among the study’s African Americans and Haitians. Qualitative data concerning participants’ knowledge of TB, gathered from 16 individual interviews further informed this analysis. Secondary research questions examined the role of spirituality, knowledge of TB and medication adherence among African Americans and Haitians. Four common themes emerged across both groups to answer the secondary research questions. Interviews revealed the themes: (a) God is in control, (b) stigmatization of TB, (c) lack of knowledge, and (d) fear of death. The theme lack of knowledge about TB was found to contribute to stigmatization of TB patients. However, in this study stigma and lack of knowledge were related to initial denial of symptoms and delayed diagnosis, but not found to be related to TB medication adherence. This study could help adult educators and health educators enhance their educational interventions, develop a better understanding of adult learning, resulting in early diagnosis and treatment ultimately decreasing transmission of TB, drug resistance, and potential death. Educators should be aware that TB patients’ spirituality may be an important part of how they cope with having TB. A larger scale study, conducted at multiple locations should be conducted to extend the findings of this small scale exploratory study. Further studies should be done to better determine what patient, healthcare provider and health care system factors might mediate relationships that may exist between lack of knowledge of TB, stigma and TB medication adherence.

A Mixed Method Study of Spirituality and the Influence on TB Medication Adherence

Nonadherence to medication for tuberculosis (TB) can lead to new cases of TB and death. Interest in spirituality in healthcare has grown among adult educators, health educators and healthcare workers (Tisdell, 2003). This mixed-method study will explore spirituality and TB medication adherence among African American and Haitian populations.