Co-saliency detection within a single image is a common vision problem that has not yet been well addressed. Existing methods often used a bottom-up strategy to infer co-saliency in an image in which salient regions a...
详细信息
Co-saliency detection within a single image is a common vision problem that has not yet been well addressed. Existing methods often used a bottom-up strategy to infer co-saliency in an image in which salient regions are firstly detected using visual primitives such as color and shape and then grouped and merged into a co-saliency map. However, co-saliency is intrinsically perceived complexly with bottom-up and top-down strategies combined in human vision. To address this problem, this study proposes a novel end-toend trainable network comprising a backbone net and two branch nets. The backbone net uses ground-truth masks as top-down guidance for saliency prediction, whereas the two branch nets construct triplet proposals for regional feature mapping and clustering, which drives the network to be bottom-up sensitive to co-salient regions. We construct a new dataset of 2019 natural images with co-saliency in each image to evaluate the proposed method. Experimental results show that the proposed method achieves state-of-the-art accuracy with a running speed of 28 fps.
Over-the-air computation(AirComp)enables federated learning(FL)to rapidly aggregate local models at the central server using waveform superposition property of wireless *** this paper,a robust transmission scheme for ...
详细信息
Over-the-air computation(AirComp)enables federated learning(FL)to rapidly aggregate local models at the central server using waveform superposition property of wireless *** this paper,a robust transmission scheme for an AirCompbased FL system with imperfect channel state information(CSI)is *** model CSI uncertainty,an expectation-based error model is *** main objective is to maximize the number of selected devices that meet mean-squared error(MSE)requirements for model broadcast and model *** problem is formulated as a combinatorial optimization problem and is solved in two ***,the priority order of devices is determined by a sparsity-inducing ***,a feasibility detection scheme is used to select the maximum number of devices to guarantee that the MSE requirements are *** alternating optimization(AO)scheme is used to transform the resulting nonconvex problem into two convex *** results illustrate the effectiveness and robustness of the proposed scheme.
This article introduces a novel approach to bolster the robustness of Deep Neural Network (DNN) models against adversarial attacks named "Targeted Adversarial Resilience Learning (TARL)". The initial ev...
详细信息
Anomaly detection(AD) has been extensively studied and applied across various scenarios in recent years. However, gaps remain between the current performance and the desired recognition accuracy required for practical...
详细信息
Anomaly detection(AD) has been extensively studied and applied across various scenarios in recent years. However, gaps remain between the current performance and the desired recognition accuracy required for practical *** paper analyzes two fundamental failure cases in the baseline AD model and identifies key reasons that limit the recognition accuracy of existing approaches. Specifically, by Case-1, we found that the main reason detrimental to current AD methods is that the inputs to the recovery model contain a large number of detailed features to be recovered, which leads to the normal/abnormal area has not/has been recovered into its original state. By Case-2, we surprisingly found that the abnormal area that cannot be recognized in image-level representations can be easily recognized in the feature-level representation. Based on the above observations, we propose a novel recover-then-discriminate(ReDi) framework for *** takes a self-generated feature map(e.g., histogram of oriented gradients) and a selected prompted image as explicit input information to address the identified in Case-1. Additionally, a feature-level discriminative network is introduced to amplify abnormal differences between the recovered and input representations. Extensive experiments on two widely used yet challenging AD datasets demonstrate that ReDi achieves state-of-the-art recognition accuracy.
The importance of secure data sharing in fog computing is increasing due to the growing number of Internet of Things(IoT)*** article addresses the privacy and security issues brought up by data sharing in the context ...
详细信息
The importance of secure data sharing in fog computing is increasing due to the growing number of Internet of Things(IoT)*** article addresses the privacy and security issues brought up by data sharing in the context of IoT fog *** suggested framework,called"BlocFogSec",secures key management and data sharing through blockchain consensus and smart *** existing solutions,BlocFogSec utilizes two types of smart contracts for secure key exchange and data sharing,while employing a consensus protocol to validate transactions and maintain blockchain *** process and store data effectively at the network edge,the framework makes use of fog computing,notably reducing latency and raising *** successfully blocks unauthorized access and data breaches by restricting transactions to authorized *** addition,the framework uses a consensus protocol to validate and add transactions to the blockchain,guaranteeing data accuracy and *** compare BlocFogSec's performance to that of other models,a number of simulations are *** simulation results indicate that BlocFogSec consistently outperforms existing models,such as Security Services for Fog Computing(SSFC)and Blockchain-based Key Management Scheme(BKMS),in terms of throughput(up to 5135 bytes per second),latency(as low as 7 ms),and resource utilization(70％to 92％).The evaluation also takes into account attack defending accuracy(up to 100％),precision(up to 100％),and recall(up to 99.6％),demonstrating BlocFogSec's effectiveness in identifying and preventing potential attacks.
In this paper the authors consider the operational problem of optimal signalling and control,called control-coding capacity(with feedback),C_(FB) in bits/second,of discrete-time nonlinear partially observable stochast...
详细信息
In this paper the authors consider the operational problem of optimal signalling and control,called control-coding capacity(with feedback),C_(FB) in bits/second,of discrete-time nonlinear partially observable stochastic systems in state space form,subject to an average cost constraint.C_(FB) is the maximum rate of encoding signals or messages into randomized controller-encoder strategies with feedback,which control the state of the system,and reproducing the messages at the output of the system using a decoder or estimator with arbitrary small asymptotic error *** the first part of the paper,the authors characterize C_(FB) by an information theoretic optimization problem of maximizing directed information from the inputs to the outputs of the system,over randomized strategies(controller-encoders).The authors derive equivalent characterizations of C_(FB),using randomized strategies generated by either uniform or arbitrary distributed random variables(RVs),sufficient statistics,and a posteriori distributions of nonlinear filtering *** the second part of the paper,the authors analyze C_(FB) for linear-quadratic Gaussian partially observable stochastic systems(LQG-POSSs).The authors show that randomized strategies consist of control,estimation and signalling parts,and the sufficient statistics are,two Kalman-filters and an orthogonal innovations *** authors prove a semi-separation principle which states,the optimal control strategy is determined explicitly from the solution of a control matrix difference Riccati equation(DRE),independently of the estimation and signalling ***,the authors express the optimization problem of C_(FB) in terms of two filtering matrix DREs,a control matrix DRE,and the covariance of the innovations *** the paper,the authors illustrate that the expression of C_(FB) includes as degenerate cases,problems of stochastic optimal control and channel capacity of information transmission.
High-dimensional and incomplete(HDI) matrices are primarily generated in all kinds of big-data-related practical applications. A latent factor analysis(LFA) model is capable of conducting efficient representation lear...
详细信息
High-dimensional and incomplete(HDI) matrices are primarily generated in all kinds of big-data-related practical applications. A latent factor analysis(LFA) model is capable of conducting efficient representation learning to an HDI matrix,whose hyper-parameter adaptation can be implemented through a particle swarm optimizer(PSO) to meet scalable ***, conventional PSO is limited by its premature issues,which leads to the accuracy loss of a resultant LFA model. To address this thorny issue, this study merges the information of each particle's state migration into its evolution process following the principle of a generalized momentum method for improving its search ability, thereby building a state-migration particle swarm optimizer(SPSO), whose theoretical convergence is rigorously proved in this study. It is then incorporated into an LFA model for implementing efficient hyper-parameter adaptation without accuracy loss. Experiments on six HDI matrices indicate that an SPSO-incorporated LFA model outperforms state-of-the-art LFA models in terms of prediction accuracy for missing data of an HDI matrix with competitive computational ***, SPSO's use ensures efficient and reliable hyper-parameter adaptation in an LFA model, thus ensuring practicality and accurate representation learning for HDI matrices.
Advancements in neuromorphic computing have given an impetus to the development of systems with adaptive behavior,dynamic responses,and energy efficiency *** charge-based or emerging memory technologies such as memris...
详细信息
Advancements in neuromorphic computing have given an impetus to the development of systems with adaptive behavior,dynamic responses,and energy efficiency *** charge-based or emerging memory technologies such as memristors have been developed to emulate synaptic plasticity,replicating the key functionality of neurons—integrating diverse presynaptic inputs to fire electrical impulses—has remained *** this study,we developed reconfigurable metal-oxide-semiconductor capacitors(MOSCaps)based on hafnium diselenide(HfSe2).The proposed devices exhibit(1)optoelectronic synaptic features and perform separate stimulus-associated learning,indicating considerable adaptive neuron emulation,(2)dual light-enabled charge-trapping and memcapacitive behavior within the same MOSCap device,whose threshold voltage and capacitance vary based on the light intensity across the visible spectrum,(3)memcapacitor volatility tuning based on the biasing conditions,enabling the transition from volatile light sensing to non-volatile optical data *** reconfigurability and multifunctionality of MOSCap were used to integrate the device into a leaky integrate-and-fire neuron model within a spiking neural network to dynamically adjust firing patterns based on light stimuli and detect exoplanets through variations in light intensity.
Vision sensors are versatile and can capture a wide range of visual cues, such as color, texture, shape, and depth. This versatility, along with the relatively inexpensive availability of machine vision cameras, playe...
详细信息
Vision sensors are versatile and can capture a wide range of visual cues, such as color, texture, shape, and depth. This versatility, along with the relatively inexpensive availability of machine vision cameras, played an important role in adopting vision-based environment perception systems in autonomous vehicles (AVs). However, vision-based perception systems can be easily affected by glare in the presence of a bright source of light, such as the sun or the headlights of the oncoming vehicle at night or simply by light reflecting off snow or ice-covered surfaces;scenarios encountered frequently during driving. In this paper, we investigate various glare reduction techniques, including the proposed saturated pixel-aware glare reduction technique for improved performance of the computer vision (CV) tasks employed by the perception layer of AVs. We evaluate these glare reduction methods based on various performance metrics of the CV algorithms used by the perception layer. Specifically, we considered object detection, object recognition, object tracking, depth estimation, and lane detection which are crucial for autonomous driving. The experimental findings validate the efficacy of the proposed glare reduction approach, showcasing enhanced performance across diverse perception tasks and remarkable resilience against varying levels of glare. IEEE
Deep reinforcement learning(DRL) has demonstrated significant potential in industrial manufacturing domains such as workshop scheduling and energy system ***, due to the model's inherent uncertainty, rigorous vali...
详细信息
Deep reinforcement learning(DRL) has demonstrated significant potential in industrial manufacturing domains such as workshop scheduling and energy system ***, due to the model's inherent uncertainty, rigorous validation is requisite for its application in real-world tasks. Specific tests may reveal inadequacies in the performance of pre-trained DRL models, while the “black-box” nature of DRL poses a challenge for testing model behavior. We propose a novel performance improvement framework based on probabilistic automata,which aims to proactively identify and correct critical vulnerabilities of DRL systems, so that the performance of DRL models in real tasks can be improved with minimal model ***, a probabilistic automaton is constructed from the historical trajectory of the DRL system by abstracting the state to generate probabilistic decision-making units(PDMUs), and a reverse breadth-first search(BFS) method is used to identify the key PDMU-action pairs that have the greatest impact on adverse outcomes. This process relies only on the state-action sequence and final result of each trajectory. Then, under the key PDMU, we search for the new action that has the greatest impact on favorable results. Finally, the key PDMU, undesirable action and new action are encapsulated as monitors to guide the DRL system to obtain more favorable results through real-time monitoring and correction mechanisms. Evaluations in two standard reinforcement learning environments and three actual job scheduling scenarios confirmed the effectiveness of the method, providing certain guarantees for the deployment of DRL models in real-world applications.
暂无评论