On weak monotonicity of some mixture functions

Monotonicity with respect to all arguments is fundamental to the definition of aggregation functions. Here we study means that are not necessarily monotone. Weak monotonicity was recently proposed as a relaxation of the monotonicity condition for averaging functions. We provide results for the weak monotonicity of some importantclasses of mixture functions. With these results we are able to extend and improve the understanding of this very important class of functions.

Composition dependency of the glass forming ability (GFA) in Mg—Ni—Si system by mechanical alloying

The pure elemental powder mixtures with the compositions of Mg₆₅Ni_xSi₃₅−_x (x = 10, 20, 25, 33 at.%) were subject to high-energy ball mill, and the structures of the mixtures at different intervals of milling were characterised by X-ray diffraction (XRD). The compositional dependency of the glass forming ability (GFA) in Mg–Ni–Si system was evaluated based on the experimental results and the theoretical calculation. The compositional dependency of GFA in Mg–Ni–Si system can be understood well by comparing the enthalpies of the crystalline and amorphous phases based on the Miedema's theory for the formation enthalpy of alloys. Increasing the Ni/Mg ratio and/or decreasing Si content can improve the amorphous formability. The calculation results might be of great help in optimising the composition with high GFA in Mg–Ni–Si system.

Investigation of nonisothermal crystallization behaviors of poly and silica nanocomposites using differential scanning calorimetry

Nonisothermal crystallization behaviors of PVA and poly (vinyl alcohol) and Silica (PVA/SiO2) nanocomposites prepared via a self-assembly monolayer (SAM) technique are investigated in this study. Differential scanning calorimetry (DSC) is used to measure the crystallization temperature and enthalpy of PVA and nanocomposites in nitrogen at various cooling rate. The results show that the degree of crystallinity of PVA and nanocomposites decreases when the SiO₂ content increases but increases with an increasing cooling rate. The peak crystallization temperature decreases with an increasing cooling rate.

Development of hybrid composites for automotive applications: effect of addition of SEBS on the morphology, mechanical, viscoelastic, crystallization and thermal degradation properties of PP/PS-xGnP composites

In this article, we report on a simple and cost effective approach for the development of light-weight, super-tough and stiff material for automotive applications. Nanocomposites based on PP/PS blend and exfoliated graphene nanoplatelets (xGnP) were prepared with and without SEBS. Mechanical, crystallization and thermal degradation properties were determined and correlated with phase morphology. The addition of xGnP to PP/PS blend increased the tensile modulus at the expense of toughness. The presence of xGnP increased the enthalpy of crystallization and enthalpy of fusion of PP in the blends, without affecting segmental mobility and thermal stability. Addition of polystyrene-block-poly(ethylene-ran-butylene)-block-polystyrene (SEBS) improved the toughness of PP/PS blends, but decreased the stiffness. The incorporation of xGnP into this ternary blend generated a super-tough material with improved stiffness and tensile elongation, suitable for automotive applications. It is observed that the presence of SEBS diminished the tendency of agglomeration of xGnP and their unfavorable interactions with thermoplastics, which in turn reduced the internal friction in the matrix.

Topic model kernel : an empirical study towards probabilistically reduced features for classification

Probabilistic topic models have become a standard in modern machine learning with wide applications in organizing and summarizing ‘documents’ in high-dimensional data such as images, videos, texts, gene expression data, and so on. Representing data by dimensional reduction of mixture proportion extracted from topic models is not only richer in semantics than bag-of-word interpretation, but also more informative for classification tasks. This paper describes the Topic Model Kernel (TMK), a high dimensional mapping for Support Vector Machine classification of data generated from probabilistic topic models. The applicability of our proposed kernel is demonstrated in several classification tasks from real world datasets. We outperform existing kernels on the distributional features and give the comparative results on non-probabilistic data types.

Topic model kernel classification with probabilistically reduced features

Probabilistic topic models have become a standard in modern machine learning to deal with a wide range of applications. Representing data by dimensional reduction of mixture proportion extracted from topic models is not only richer in semantics interpretation, but could also be informative for classification tasks. In this paper, we describe the Topic Model Kernel (TMK), a topicbased kernel for Support Vector Machine classification on data being processed by probabilistic topic models. The applicability of our proposed kernel is demonstrated in several classification tasks with real world datasets. TMK outperforms existing kernels on the distributional features and give comparative results on nonprobabilistic data types.

Stereocomplexation and monolayer morphologies of a stereoregular poly(methylmethacrylate) mixture formed at the air/water surface

The stereocomplexation of stereoregular PMMA at the air/water interface was proved by structure determination using reflection-absorption IR and grazing incidence X-ray diffraction. Morphological studies on LB films of i- and s-PMMA blends with different ratios help to disclose the stereocomplexation process of stereoregular PMMA at the air/water interface. It was found that the stereocomplexes exist in particle aggregates randomly dispersed at the air/water interface. In the systems with the i:s ratio deviated from 1:2, the molecules, either i-PMMA or s-PMMA, that do not participate in the stereocomplexation build separate layer surrounding the stereocomplexes. This layer is much thinner than the particle aggregates of the stereocomplexes. If the i-PMMA molecules are rich in this thinner layer, crystallization of i-PMMA takes place, which generates lamellar structure besides the stereocomplexes.

Influence of N2O flow rate on reliability of SiOx films deposited by SiH4-N2O gas mixture plasma

SiO_x films have several advantages as an interlayer dielectric in electronic devices owing to the strong adhesion between SiO_x and the substrate. In this study, the coating performance as a function of the N₂O flow rate was evaluated by electrochemical impedance spectroscopy and potentiodynamic polarization tests in an undisturbed environment. In addition, the coatings were examined by atomic force microscopy and Fourier transform infrared reflection spectroscopy. The SiO_x films on a stainless-steel substrate showed the highest coating performance at a N₂O flow rate of 120 sccm. This was attributed to the films having the lowest porosity value among those examined as a result of the fragmentation of SiO and SiO₂ bonds and the improved surface roughness.

Incremental learning of temporally-coherent gaussian mixture models

In this paper we address the problem of learning Gaussian Mixture Models (GMMs) incrementally. Unlike previous approaches which universally assume that new data comes in blocks representable by GMMs which are then merged with the current model estimate, our method works for the case when novel data points arrive oneby- one, while requiring little additional memory. We keep only two GMMs in the memory and no historical data. The current fit is updated with the assumption that the number of components is fixed, which is increased (or reduced) when enough evidence for a new component is seen. This is deduced from the change from the oldest fit of the same complexity, termed the Historical GMM, the concept of which is central to our method. The performance of the proposed method is demonstrated qualitatively and quantitatively on several synthetic data sets and video sequences of faces acquired in realistic imaging conditions