We propose a new variational inference method which uses recognition models for amortized inference in graphical models that contain deep generative models. Unlike many existing approaches, our method can handle non-conjugacy in both the latent graphical model and the deep generative model, and enables fully amortized inference at test time. Our method is based on an extension of a recently pro...

This paper describes a multimodal dialogue driven system, ARCHIVUS, that allows users to access and retrieve the content of recorded and annotated multimodal meetings. We describe (1) a novel approach taken in designing the system given the relative inapplicability of standard user requirements elicitation methodologies, (2) the components of ARCHIVUS, and (3) the methodologies that we plan to ...

Simulation is an eeective technique to test software where testing in the real environment is not feasible. A 3D graphical simulator can make the software testing process much more eeective and eecient. With graphical representation, the softwares behavior can be observed in a virtual environment similar to the real environment. This kind of simulator can be best used in testing embedded system...

graphics display is one of the most useful methods of interacting with a system graphically. This paper examines existing graphical support and lists requirements for high-level support of graphical interaction. The architecture of a prototype system with high-level support for graphical interaction is presented. This includes database support for manipulating graphical data and device-independ...

CAISARTS (Conceptual, Analytical, and Implementation Scheduling Advice for Real-Time Systems) is a rule-based system used by real-time application designers to obtain expert assistance for all aspects of the design related to scheduling: granularity of tasks, allocation of tasks, choice and analysis of scheduling paradigm, analysis of overheads of particular operating systems and scheduling par...

Environmental pollution has caused governments to be concerned about energy saving and the reduction of environmental pollution. Some researchers have presented resource allocation models as multi-objective linear programming (MOLP) in order pay more attention reduction. Energy affects both desirable undesirable outputs. In this paper, we argue for inapplicability existing reducing outputs thro...

‎A natural algorithm with an optimal order of convergence is proposed for numerical solution of a class of cordial weakly singular Volterra integral equations‎. ‎The equations of this class appear in heat conduction problems with mixed boundary conditions‎. ‎The algorithm is based on a representation of the solution and compound Gaussian quadrature rules with graded meshes‎. ‎A comparative stud...

This manuscript addresses the creation of scenario-based models to reason about the behavior of existing industrial information systems. In our approach the system behavior is modeled in two steps that gradually introduce detail and formality. This manuscript addresses the first step, where text-based descriptions, in the form of structured rules, are used to specify how the system is or should...