The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The deployments of deep neural network models on mobile or embedded devices have been challenged due to two main reasons: 1) the large model size for storage, and 2) the large memory bandwidth for inference. To address these issues, this paper develops a deep neural network compression framework to reduce the resource usage for efficient visual inference. By reviewing the trained deep model, we propose...
In this paper, a dynamic offloading model is proposed to minimize the energy consumption of mobile devices by exploiting cloud computational resources for view synthesis. The computational complexity of view synthesis, the processing capability of the cloud, the processing capability and the power consumption of the mobile are considered jointly into the model to provide an optimized solution. Several...
As one kind with the best comprehensive performance of various honeycombs, triangular honeycomb is an important structure to achieve lightweight. Focusing on the deficiency of the non-optimized macro-material distribution in uniform triangular honeycomb, a design and modeling method of variable-density triangular honeycomb structures was proposed. With a three-point bending beam as the design object,...
Several conventional methods have been implemented in pattern recognition, but few of them have biological plausibility. This paper mimics the hierarchical visual system and uses the precise-spike-driven (PSD) synaptic plasticity rule to learn. The well-known HMAX model imitates the visual cortex and uses Gabor filter and max pooling to extract features. Compared with the traditional HMAX model, our...
Cloud Infrastructure as a Service (IaaS), where traditional IT infrastructure resources such as compute, storage and networking are owned by a cloud service provider (CSP) and offered as on-demand virtual resources to customers (tenants), is the fastest maturing service model in cloud computing. The transformation of physical resources into virtual offers great flexibility to CSP customers including...
Visualization technology helps people find much information which is difficult to dig up with the conventional statistical methods, understand the dynamic variation law, so as to predict the development and the results of network better. In this paper, based on the relevant knowledge of complex network theory and communication, we use the huge amounts of data in the social network to set up communication...
3D-HEVC is the latest 3D video project of MPEG for 3D video coding. To compress the depth videos in 3D-HEVC efficiently, depth distortion models considering the distortions in synthesized views have been introduced. However, these models introduce a big burden in coding complexity. In this paper, a configuration of depth distortion models applied at each encoding decision stage of 3D-HEVC is proposed...
A new acoustic model based on deep neural network (DNN) has been introduced recently and outperforms the conventional Gaussian mixture model (GMM) in speech recognition on several tasks. However, the number of parameters required by a DNN model is much larger than that of its counterpart. The excessive cost of computation cumbers the implementation of DNN in many scenarios. In this paper, a DNN-based...
This paper discusses the NP-complete multi-constrained path problem with imprecise state information, and a heuristic pre-computation algorithm based on flooding with imprecise additive link state information is presented. This algorithm uses limited flooding policy based on selective non-dominated path and can find paths from present node to all other nodes only running once. And the algorithm deals...
According to the distributed substation protection configuration and the principle of fault-clearance, a new model of fault diagnosis based on Petri net is proposed in this paper. In Petri net diagnosis model, all kinds of fault have a specific token which make it easily and clearly to find the fault location and understand the sequence of the fault events. The diagnostic is implemented by solving...
This paper presents the algorithm and software developed for parallel simulation of spiking neural networks on multiple SpiNNaker universal neuromorphic chips. It not only describes approaches to simulating neural network models, such as dynamics, neural representations, and synaptic delays, but also presents the software design of loading a neural application and initial a simulation on the multi-chip...
SpiNNaker is a massively parallel architecture with more than a million processing cores that can model up to 1 billion spiking neurons in biological real time. Here, we offer an overview of our research project and describe the first experiments with these test chips running spiking neurons based on Eugene Izhikevich's model. Note that we're not targeting artificial neural networks (such as perceptrons...
At present, the simulation model does not fully reflect the effects of machining errors, surface quality and assembly errors which are produced during the manufacturing process on the performance of parts. This makes the simulation results differ greatly from experiment results. In order to solve this problem, simulation modeling technology of the performance of mechanical system based on manufacturing...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.