A5055361145- Neural Networks and Applications
- Neural Networks Stability and Synchronization
- Radiation Effects in Electronics
- VLSI and Analog Circuit Testing
- Complex Network Analysis Techniques
- Opinion Dynamics and Social Influence
- Advanced Graph Neural Networks
- Image Processing Techniques and Applications
- Metaheuristic Optimization Algorithms Research
- Advanced Memory and Neural Computing
- Thermography and Photoacoustic Techniques
- Advancements in Semiconductor Devices and Circuit Design
- Face recognition and analysis
- Image and Signal Denoising Methods
- Photoacoustic and Ultrasonic Imaging
- Low-power high-performance VLSI design
- Recommender Systems and Techniques
- Traffic and Road Safety
- Integrated Circuits and Semiconductor Failure Analysis
- Aerospace Engineering and Energy Systems
- Advanced Image and Video Retrieval Techniques
- Wind Energy Research and Development
- Stress Responses and Cortisol
- Advanced Clustering Algorithms Research
- Neural Networks and Reservoir Computing
Zhejiang University of Technology
2015-2025
Jiangxi University of Technology
2017
Northeastern University
2010-2013
IQ Samhällsbyggnad
2008
State University of New York
1989
In biology and medicine research, detection identification of cancer cells plays an essential role to further analysis cell properties developing new drugs experiments. However, owing the adhesion among great changes in morphology, it is a very challenging task detect locate accurately, especially for area. this work, deep detector based on framework Faster R-CNN proposed, this, Circle Scanning Algorithm (CSA) presented redetection cells. And then series experiments are achieved. The results...
Currently, few works focus on the reliability of blockchain architecture at circuit level, which makes its security and privacy vulnerable to hardware errors. To mitigate hazard, through an iterative probabilistic transfer matrix model coded by a binary-decimal mechanism, reliability-based evaluation method for importance units is proposed in this paper. First, calculates output reliabilities any leads obtains gradient based barrel theory. Second, it sorts their obtained gradients. Third,...
The increase in the reliability requirements of integrated circuits applied diverse smart sensing devices and cost test generation fault simulation have expanded need for new approaches to estimate signal logic circuits, which will help trust management Internet Things systems. This paper presents a novel method analysis with unreliable application trust-driven design. Based on extended probabilistic transfer matrix model binary-decimal coding allocation, by using technologies state-vector...
Hardening reliability-critical gates in a circuit is an important step to improve the reliability at low cost. However, accurately locating key prerequisite for efficient implementation of hardening operation. In this paper, probabilistic-based calculation method developed described. The proposed based on generation input vectors and sampling using uniform non-Bernoulli sequences, criticality gate measured by combining structure information itself. Both accuracy efficiency have been...
Abstract Aggression is a common and complex social behavior that associated with violence mental diseases. Although sex differences were observed in aggression, the neural mechanism for effect of on aggression behaviors remains unclear, especially specific subscales aggression. In this study, we investigated effects subscales, gray matter volume (GMV), functional connectivity (FC) each insula subregion as well correlation GMV FC. This study found significantly influenced (a) physical anger,...
The link prediction algorithm is one of the key technologies to reveal inherent rule network evolution. This paper proposes a novel based on properties local community, which composed common neighbor nodes any two in and links between these nodes. By referring node degree condition assortativity or disassortativity network, we comprehensively consider effect shortest path edge clustering coefficient within community similarity. We numerically show proposed method provide good results.
The multistability problem is discussed for 2n-dimensional bidirectional associative memory (BAM) neural networks with a general class of activation functions. By using analysis approach and decomposition state space, ¿ <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2n</sup> can be divided into 3 xmlns:xlink="http://www.w3.org/1999/xlink">n</sup> regions. 2 regions them are positive invariant, under some conditions, every invariant set has...
In this paper, the multistability issue is discussed for delayed complex-valued recurrent neural networks with discontinuous real-imaginary-type activation functions. Based on a fixed theorem and stability definition, sufficient criteria are established existence of multiple equilibria networks. The number stable larger than that real-valued networks, which can be used to achieve high-capacity associative memories. One numerical example provided show effectiveness superiority presented results.
Reliable circuits help prevent artificial intelligence (AI) systems from being corrupted by the soft errors occurred in memories or combinational circuits, which promotes development of AI security. However, it is a great challenge to measure reliability at register transfer level (RTL) rapidly and efficiently. In this paper, new fast accurate computational model based on stochastic computation (SC) presented meet these objectives. proposed approach, circuit netlists RTL are parsed satisfy...
Vascular centerlines have crucial significance in reconstruction, registration, segmentation and vascular parameter analysis. The extraction of vessel structures remains a difficult problem the completeness continuity results. In this paper, we present novel method to extract cerebrovascular from four-dimensional computed tomography angiography images. Tubular features directions are used initial centerlines, offset correction is introduced orthogonal plane. addition, also post-processing...
Multiple stability for two-dimensional delayed recurrent neural networks with piecewise linear activation functions of 2r (r ≥ 1) corner points is studied. Sufficient conditions are established checking the existence (2r + 1)2 equilibria in networks. Under these conditions, locally exponentially stable, and — r2 unstable. Attractive basins stable estimated, which larger than invariant sets derived by decomposing state space. One example provided to illustrate effectiveness our results.
In developing the normal equation of finite-duration impulse-response Wiener filter, method taking a gradient scalar-valued mean-square error to complex tap-weight vector is discussed. A variational used attain also included.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>