Deep neural networks (DNNs) have demonstrated exceptional performance, leading to diverse applications across various mobile devices (MDs). Considering factors like portability and environmental sustainability, an inc...
详细信息
ISBN:
(数字)9783982674100
ISBN:
(纸本)9798331534646
Deep neural networks (DNNs) have demonstrated exceptional performance, leading to diverse applications across various mobile devices (MDs). Considering factors like portability and environmental sustainability, an increasing number of MDs are adopting energy harvesting (EH) techniques for power supply. However, the computational intensity of DNNs presents significant challenges for their deployment on these resource-constrained devices. Existing approaches often employ DNN partition or offloading to mitigate the time and energy consumption associated with running DNNs on MDs. Nonetheless, existing methods frequently fall short in accurately modeling the execution time of DNNs, and do not consider to use thread allocation for further latency and energy consumption optimization. To solve these problems, we propose a dynamic DNN partition and thread allocation method to optimize the latency and energy consumption of running DNNs on EH-enabled MDs. Specifically, we first investigate the relationship between DNN inference latency and allocated threads and establish an accurate DNN latency prediction model. Based on the prediction model, a DRL-based DNN partition (DDP) algorithm is designed to find the optimal partitions for DNNs. A thread allocation (TA) algorithm is proposed to reduce the inference latency. Experimental results from our test-bed platform demonstrate that compared to four benchmarking methods, our scheme can reduce DNN inference latency and energy consumption by up to 37.3% and 38.5%.
Anomalies are usually regarded as data errors or novel patterns previously unseen, which are quite different from most observed data. Accurate detection of anomalies is crucial in various application scenarios. This p...
详细信息
Existing visual saliency prediction methods mainly focus on single-modal visual saliency prediction, while ignoring the significant impact of text on visual saliency. To more comprehensively explore the influence of t...
详细信息
Recent years have witnessed rapid progress of convolutional neural networks (CNNs) and their successful application in the task of saliency prediction for omnidirectional images (ODIs). Albeit achieving tremendous per...
详细信息
In this paper, the problem of remote state estimation is investigated for a class of complex networks with noisy wireless communication channels. The employment of the binary encoding scheme allows for the description...
详细信息
In this paper,we study the motion course of traffic flow on the slopes of a highway by applying a microscopic traffic model,which takes into account the next-nearest-neighbor interaction in an intelligent transportati...
详细信息
In this paper,we study the motion course of traffic flow on the slopes of a highway by applying a microscopic traffic model,which takes into account the next-nearest-neighbor interaction in an intelligent transportation system *** common gradients of the highway,which are sag terrain,uphill terrain,and downhill terrain on a single-lane roadway,are selected to clarify the impact on the traffic flow by the next-nearest-neighbor interaction in relative *** obtain the current-density relation for traffic flow on the sag,the uphill and the downhill under the next-nearest-neighbor interaction *** is observed that the current saturates when the density is greater than a critical value and the current decreases when the density is greater than another critical *** the density falls into the intermediate range between the two critical densities it is also found that the oscillatory jam,easily leads to traffic accidents,often appears in the downhill stage,and the next-nearest-neighbor interaction in relative velocity has a strong suppressing effect on this kind of dangerous congestion.A theoretical analysis is also presented to explain this important conclusion.
In order to guarantee the correctness of business processes, not only control-flow errors but also data-flow errors should be considered. The control-flow errors mainly focus on deadlock, livelock, soundness, and so o...
详细信息
In order to guarantee the correctness of business processes, not only control-flow errors but also data-flow errors should be considered. The control-flow errors mainly focus on deadlock, livelock, soundness, and so on. However, there are not too many methods for detecting data-flow errors. This paper defines Petri nets with data operations(PN-DO) that can model the operations on data such as read, write and delete. Based on PN-DO, we define some data-flow errors in this paper. We construct a reachability graph with data operations for each PN-DO, and then propose a method to reduce the reachability graph. Based on the reduced reachability graph, data-flow errors can be detected rapidly. A case study is given to illustrate the effectiveness of our methods.
Inter-organizational workflow nets(IWF-nets) can well model the interactions among multiple processes by sending/receiving messages. Compatibility and weak compatibility are crucial properties for *** latter guarantee...
详细信息
Inter-organizational workflow nets(IWF-nets) can well model the interactions among multiple processes by sending/receiving messages. Compatibility and weak compatibility are crucial properties for *** latter guarantees that a system is deadlock-free and livelock-free while the former also guarantees that it has no dead tasks. Our previous work proved that the(weak) compatibility problem is PSPACE-complete for safe IWF-nets. This paper defines a class of IWF-nets in which some simple circuits are allowed. Necessary and sufficient conditions are presented to decide compatibility and weak compatibility for this class, and they are dependent on the net structures only. Algorithms are developed based on these conditions. In addition, we show that the traditional net structures like siphon cannot be easily used to decide the(weak) compatibility of IWF-nets.
The rapid development of location-based social networks(LBSNs) provides people with an opportunity of better understanding their mobility behavior which enables them to decide their next *** example,it can help travel...
详细信息
The rapid development of location-based social networks(LBSNs) provides people with an opportunity of better understanding their mobility behavior which enables them to decide their next *** example,it can help travelers to choose where to go next,or recommend salesmen the most potential places to deliver advertisements or sell *** this paper,a method for recommending points of interest(POIs)is proposed based on a collaborative tensor factorization(CTF)***,a generalized objective function is constructed for collaboratively factorizing a tensor with several feature ***,a 3-mode tensor is used to model all users' check-in behaviors,and three feature matrices are extracted to characterize the time distribution,category distribution and POI correlation,***,each user's preference to a POI at a specific time can be estimated by using *** order to further improve the recommendation accuracy,PCTF(Partitionbased CTF) is proposed to fill the missing entries of a tensor after clustering its every *** on a real checkin database show that the proposed method can provide more accurate location recommendation.
As a powerful analysis tool of Petri nets, reachability trees are fundamental for systematically investigating many characteristics such as boundedness, liveness and reversibility. This work proposes a method to gener...
详细信息
暂无评论