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.
Deep Neural Networks (DNNs) are pervasively used in a significant number of applications and platforms. To enhance the execution efficiency of large-scale DNNs, previous attempts focus mainly on client-server paradigms, relying on powerful external infrastructure, or model compression, with complicated pre-processing phases. Though effective, these methods overlook the optimization of DNNs on distributed...
Very large-scale Deep Neural Networks (DNNs) have achieved remarkable successes in a large variety of computer vision tasks. However, the high computation intensity of DNNs makes it challenging to deploy these models on resource-limited systems. Some studies used low-rank approaches that approximate the filters by low-rank basis to accelerate the testing. Those works directly decomposed the pre-trained...
In this paper, we propose a wireless battery charging system which the power transfer position can be automatically moved to the optimal charging placement. The proposed charging system is based on the electromagnetic inductive power transfer technique which is a short distance power transfer system. As a result, we have successfully finished the implementation of the proposed wireless battery charging...
Intelligent transportation systems (ITS) are becoming more and more effective, benefiting from big data. Despite this, missing data is a problem that prevents many prediction algorithms in ITS from working effectively. Much work has been done to impute those missing data. Among different imputation methods, k-nearest neighbours (kNN) has shown excellent accuracy and efficiency. However, the general...
This paper proposes an in-vehicle power line communication (PLC) system which provides an in-vehicle communication for the video and vehicle messages transmission as a reliable alternative networking medium. The proposed in-vehicle PLC system adopts an existing in-vehicle 12-V battery power system, which means we don't need an external DC power supply for in-vehicle communication equipment to achieve...
Biomechanical properties of the extracellular matrix (ECM) are important regulators in cell development, including proliferation, apoptosis and migration. Recent studies have shown that stiffening of the ECM resulting from deposition and crosslinking of collagen may promote invasion and migration of tumor cells. In our previous study, a laser speckle contrast shear wave (SW) imaging system was developed,...
Mechanobiology is an emerging field of research that studies the influences on the cell behaviors by mechanical stimuli. In this research, the previously developed laser speckle contrast shear wave imaging system is implemented for monitoring the temporal and spatial changes in the stiffness of the self-designed in vitro cancer metastasis model. Results from the immunostaining with laminin and collagen...
In this work, we propose a regularized learning method that is able to take into account the deviation of the memristor-mapped synaptic weights from the target values determined during the training process. Experimental results obtained when utilizing the MNIST data set show that compared to the conventional learning method which considers the learning and mapping processes separately, our learning...
Recently, DNN model compression based on network architecture design, e.g., SqueezeNet, attracted a lot attention. No accuracy drop on image classification is observed on these extremely compact networks, compared to well-known models. An emerging question, however, is whether these model compression techniques hurt DNNs learning ability other than classifying images on a single dataset. Our preliminary...
Evolutionary multitasking aims to explore implicit synergy among multiple optimization tasks. Through the effect of hitchhiking, evolutionary multitasking is capable of improving the performance of evolutionary algorithms on exploration as well as exploitation. Multifactorial evolutionary algorithm (MFEA) presented an effectual implementation of evolutionary multitasking, which simultaneously seeks...
In this paper, a relay selection method based on the minimum opportunity cost (MOC) is proposed to prolong the lifetime of amplify-and-forward (AF) cooperative networks with multiple radio frequency energy-harvesting (RFEH) relays, where lifetime is defined as the longest period in which outage probability of the network is lower than a predefined threshold. Compared with existing relay selection...
All active sessions of an ordinary host will be broken if the host changes its IP address as a result of migrating to a new subnet. Traditional solutions toward this problem either need modifying mobile hosts or create tunnels that cause inefficient triangle routing. SDN-based mobility schemes, on the other hand, focus on handover latency reduction or fast packet redirection after handover. There...
With the increasing popularity of light emitting diodes (LEDs), visible light communication (VLC) systems provide an attractive alternative method for wireless communication. To reduce complexity and achieve lower cost, the on-off keying (OOK) modulation has been widely adopted in implementation of VLC systems. Although the conventional minimum-voltage detection circuit can remove the interference...
Convolutional neural networks (CNNs) have recently broken many performance records in image recognition and object detection problems. The success of CNNs, to a great extent, is enabled by the fast scaling-up of the networks that learn from a huge volume of data. The deployment of big CNN models can be both computation-intensive and memory-intensive, leaving severe challenges to hardware implementations...
As a large-scale commercial spiking-based neuromorphic computing platform, IBM TrueNorth processor received tremendous attentions in society. However, one of the known issues in TrueNorth design is the limited precision of synaptic weights. The current workaround is running multiple neural network copies in which the average value of each synaptic weight is close to that in the original network. We...
Controller Area Network (CAN) has become the most popular protocol for real-time control and automation systems due to its high reliability and low cost. However, CAN suffers from a limitation on the maximum length of a single bus arising from its specified bus topology. In order to solve this drawback indicated, a new internetworking device had been proposed to divide large-scale CAN networks into...
Synapse crossbar is an elementary structure in neuromorphic computing systems (NCS). However, the limited size of crossbars and heavy routing congestion impede the NCS implementation of large neural networks. In this paper, we propose a two-step framework (namely, group scissor) to scale NCS designs to large neural networks. The first step rank clipping integrates low-rank approximation into the training...
Brain inspired neuromorphic computing has demonstrated remarkable advantages over traditional von Neumann architecture for its high energy efficiency and parallel data processing. However, the limited resolution of synaptic weights degrades system accuracy and thus impedes the use of neuromorphic systems. In this work, we propose three orthogonal methods to learn synapses with one-level precision,...
Configuration bugs are among the dominant causes of software failures. Software organizations often use bug tracking systems to manage bug reports collected from developers and users. In order for software developers to understand and reproduce configuration bugs, it is vital for them to know whether a bug in the bug report is related to configuration issues, this is not often easily discerned due...
The purpose of this research is to provide a puzzle-based framework to study how global and local information is interacted on human's visual perception during the decision making process. Since the Deep Convolutional Neural Networks (DCNN) has shown the state of the art performance in image classification and object detection, DCNN can output scores to reflect the level of global information, which...
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.