Deep neural networks have been successfully applied in applications with a large amount of labeled data. However, there are major drawbacks of the neural networks that are related to rapid generalization with small data and continual learning of new concepts without forgetting. We present a novel meta learning method, Meta Networks (MetaNet), that acquires a meta-level knowledge across tasks an...

As an alternative to the rather intensive computations when using the nonlinear traffic flow model METANET in a model-based predictive control context, a piecewise-affine (PWA) approximation of the model is proposed. Here, several model equations amongst which the fundamental diagram are approximated by a PWA function. Some selected methods to determine this approximation are shortly discussed....

MetaNet is a network management system that provides end-to-end communication services with QoS and communication planning support for mission critical applications. It supports immediate and advance resource reservation, and provides information on immediate and future available communication capacity taking into account network configuration and traffic. MetaNet supports these services over m...

In a diversified internet, meta-networks (“metanets?for short) share a common substrate and offer value-added services to millions of users around the globe. Therefore, configuring low-cost metanets with links having enough bandwidth to accommodate all anticipated user traffic is critical to the success of the metanets. In this paper, we propose a novel pruning algorithm that configures metanet...

− This work investigates quality of service (QoS) attributes at five levels in a version of an information dissemination system called the Agile Information Control Environment (AICE). These levels are: application, information flow, infochannel, MetaNet, and network. The user is interested in application QoS. An information flow is a communication need between a source-sink pair in response to...

In a diversified internet, meta-networks (“metanets?for short) share a common substrate and offer value-added services to millions of users around the globe. Therefore, configuring low-cost metanets with links having enough bandwidth to accommodate all anticipated user traffic is critical to the success of the metanets. In this paper, we propose a novel pruning algorithm that configures metanet...

We present a framework to integrate macroscopic traffic flow models with microscopic emission and fuel consumption models. Since macroscopic traffic flow models do not provide the acceleration as an output, while microscopic emission and fuel consumption models require both the instantaneous speeds and accelerations as input, we describe how to generate these variables. In particular, we consid...

Metadata interoperability is a fundamental requirement for access to information within networked knowledge organization systems. The Harmony International Digital Library Project [1] has developed a common underlying data model (the ABC model) to enable the scalable mapping of metadata descriptions across domains and media types. The ABC model, described in [2], provides a set of basic buildin...

Curated gene sets from databases such as KEGG Pathway and Gene Ontology are often used to systematically organize lists of genes or proteins derived from high-throughput data. However, the information content inherent to some relationships between the interrogated gene sets, such as pathway crosstalk, is often underutilized. A gene set network, where nodes representing individual gene sets such...