Bayesian estimation of Dirichlet mixture model with variational inference

In statistical modeling, parameter estimation is an essential and challengeable task. Estimation of the parameters in the Dirichlet mixture model (DMM) is analytically intractable, due to the integral expressions of the gamma function and its corresponding derivatives. We introduce a Bayesian estimation strategy to estimate the posterior distribution of the parameters in DMM. By assuming the gamma distribution as the prior to each parameter, we approximate both the prior and the posterior distribution of the parameters with a product of several mutually independent gamma distributions. The extended factorized approximation method is applied to introduce a single lower-bound to the variational objective function and an analytically tractable estimation solution is derived. Moreover, there is only one function that is maximized during iterations and, therefore, the convergence of the proposed algorithm is theoretically guaranteed. With synthesized data, the proposed method shows the advantages over the EM-based method and the previously proposed Bayesian estimation method. With two important multimedia signal processing applications, the good performance of the proposed Bayesian estimation method is demonstrated.     

Organic Nanoparticles with Polypropyleneoxide Chains as Support for Metallocene Catalysts: Ethylene Homopolymerization and Ethylene/α-olefin Copolymerization

Summary Novel organic nanoparticles functionalized with nucleophilic polypropyleneoxide (PPO) chains on their surfaces for supporting metallocene catalysts in heterogeneous olefin polymerization are presented. The nanoparticles (60–100 nm) were obtained by miniemulsion polymerization of styrene, divinylbenzene and PPO functionalized styrene. It is demonstrated that Me₂Si(2MeBenzlnd)₂ZrCl₂/MAO supported on these nanoparticles is suitable for the homopolymerization of ethylene, resulting in excellent product morphologies and high activities. lt is shown that by varying the MAO/Zr ratios and Zr concentrations the activities and productivities of the catalysts as well as the qualities of the polyethylene products can be tuned. These new supported catalysts are also suitable for the copolymerization of ethylene with several comonomers (1-hexene, 1-octene, 1-decene or norbornene). As the obtained product properties like crystallinity, melting temperature or bulk density match the results of silica supported systems, these organic nanoparticles can be considered as alternative carriers in comparison to the established inorganic ones.     

Branching-time logics with path relativisation

We define extensions of the full branching-time temporal logic CTL^? in which the path quantifiers are relativised by formal languages of infinite words, and consider its natural fragments obtained by extending the logics CTL and CTL⁺ in the same way. This yields a small and two-dimensional hierarchy of temporal logics parametrised by the class of languages used for the path restriction on one hand, and the use of temporal operators on the other. We motivate the study of such logics through two application scenarios: in abstraction and refinement they offer more precise means for the exclusion of spurious traces; and they may be useful in software synthesis where decidable logics without the finite model property are required. We study the relative expressive power of these logics as well as the complexities of their satisfiability and model-checking problems.     

Spatio‐temporal geometry fusion for multiple hybrid cameras using moving least squares surfaces

Multi‐view reconstruction aims at computing the geometry of a scene observed by a set of cameras. Accurate 3D reconstruction of dynamic scenes is a key component for a large variety of applications, ranging from special effects to telepresence and medical imaging. In this paper we propose a method based on Moving Least Squares surfaces which robustly and efficiently reconstructs dynamic scenes captured by a calibrated set of hybrid color+depth cameras. Our reconstruction provides spatio‐temporal consistency and seamlessly fuses color and geometric information. We illustrate our approach on a variety of real sequences and demonstrate that it favorably compares to state‐of‐the‐art methods.     

On‐the‐fly generation and rendering of infinite cities on the GPU

In this paper, we present a new approach for shape‐grammar‐based generation and rendering of huge cities in real‐time on the graphics processing unit (GPU). Traditional approaches rely on evaluating a shape grammar and storing the geometry produced as a preprocessing step. During rendering, the pregenerated data is then streamed to the GPU. By interweaving generation and rendering, we overcome the problems and limitations of streaming pregenerated data. Using our methods of visibility pruning and adaptive level of detail, we are able to dynamically generate only the geometry needed to render the current view in real‐time directly on the GPU. We also present a robust and efficient way to dynamically update a scene's derivation tree and geometry, enabling us to exploit frame‐to‐frame coherence. Our combined generation and rendering is significantly faster than all previous work. For detailed scenes, we are capable of generating geometry more rapidly than even just copying pregenerated data from main memory, enabling us to render cities with thousands of buildings at up to 100 frames per second, even with the camera moving at supersonic speed.     

IISPH‐FLIP for incompressible fluids

We propose to use Implicit Incompressible Smoothed Particle Hydrodynamics (IISPH) for pressure projection and boundary handling in Fluid‐Implicit‐Particle (FLIP) solvers for the simulation of incompressible fluids. This novel combination addresses two issues of existing SPH and FLIP solvers, namely mass preservation in FLIP and efficiency and memory consumption in SPH. First, the SPH component enables the simulation of incompressible fluids with perfect mass preservation. Second, the FLIP component efficiently enriches the SPH component with detail that is comparable to a standard SPH simulation with the same number of particles, while improving the performance by a factor of 7 and significantly reducing the memory consumption. We demonstrate that the proposed IISPH‐FLIP solver can simulate incompressible fluids with a quantifiable, imperceptible density deviation below 0.1%. We show large‐scale scenarios with up to 160 million particles that have been processed on a single desktop PC using only 15GB of memory. One‐ and two‐way coupled solids are illustrated.     

Foresight for a responsible future with ICT

The ACM Code of Ethics asserts that computing professionals have an ethical responsibility to minimize the negative consequences of information and communication technologies (ICT). Negative consequences are rarely intended, but they can often be foreseen with careful sociotechnical analysis in advance of system building. Motivated by an interest in extremely complex sociotechnical contexts (e.g., mortgage lending and automated trading) where ICT appears to be having negative consequences in addition to many benefits, this paper identifies and evaluates future-oriented problem analysis and solution design tools in three potentially relevant literatures: 1) ICT ethics, 2) environmental sustainability, and 3) technology hazards. Several promising future-oriented technology analysis techniques (e.g., anticipatory technology ethics, technology roadmapping, morphological analysis, and control structure analysis) were found and are discussed in this paper, but much work remains to be done to customize them, integrate them, and codify them for use in education and high-quality IS research on very complex sociotechnical contexts like the global financial network.     

Engineering microfluidic papers: effect of fiber source and paper sheet properties on capillary-driven fluid flow

In the present study, we introduce a novel approach to control and modulate fluid transport inside microfluidic papers using lab-engineered paper sheets. Lab-sheets consisting of different fiber sources (eucalyptus sulfate and cotton linters pulp) and varying porosities were designed and further modified with small millimeter-scaled channels using hydrophobic barriers consisting of fiber-attached, hydrophobic polymers. The capillary-driven transport of an aqueous solution was monitored visually, and the influence of parameters such as fiber source, paper grammage, and channel width on the flow rates through the channel was investigated. The experimental results were compared with those obtained with commercially available filter papers. Our findings suggest that accurate control of fluid transport processes with standard filter papers is complex. Additionally, if the channel width is smaller than the mean fiber length, flow rates become dependent on the geometric parameters of the channel because of the formation of dead-end pores at the hydrophobic barriers. Finally, control of the paper sheets porosity, by varying the fiber density of the lab-made paper, affords the fabrication of chemically identical sheets whereby capillary flow is largely influenced and can be modulated accordingly by simple papermaking processes.     

An architecture modeling framework for probabilistic prediction

In the design phase of business and IT system development, it is desirable to predict the properties of the system-to-be. A number of formalisms to assess qualities such as performance, reliability and security have therefore previously been proposed. However, existing prediction systems do not allow the modeler to express uncertainty with respect to the design of the considered system. Yet, in contemporary business, the high rate of change in the environment leads to uncertainties about present and future characteristics of the system, so significant that ignoring them becomes problematic. In this paper, we propose a formalism, the Predictive, Probabilistic Architecture Modeling Framework (P²AMF), capable of advanced and probabilistically sound reasoning about business and IT architecture models, given in the form of Unified Modeling Language class and object diagrams. The proposed formalism is based on the Object Constraint Language (OCL). To OCL, P²AMF adds a probabilistic inference mechanism. The paper introduces P²AMF, describes its use for system property prediction and assessment and proposes an algorithm for probabilistic inference.     

Fluidized Bed Air Drying: Experimental Study and Model Development

The presented study describes the processes and mechanisms of batch fluidized bed drying. The influencing factors of hot air drying are theoretically and experimentally examined, in order to present the relations between temperature and humidity profiles and all other drying parameters. A physical model is presented to facilitate the calculation of the drying processes under defined conditions. Three succeeding drying stages are therefore modeled. Mass and energy balances including all components taking part in the process are formulated. The model clarities the drying process under the assumption of pure heat transfer mechanisms. It does not contain adaptive parameters and takes into account an inactive bypass fraction of the fluidization and drying medium. The evaluation of the model was successful for two fluidized bed plants with nominal widths of 100 mm and 400 mm. The experiments showed sufficient accuracy and transferability of the model to equipment of application‐oriented dimensions.