A5100401934- Anomaly Detection Techniques and Applications
- Network Security and Intrusion Detection
- Quantum Computing Algorithms and Architecture
- Power Transformer Diagnostics and Insulation
- Time Series Analysis and Forecasting
- Speech and Audio Processing
- Quantum Information and Cryptography
- Machine Fault Diagnosis Techniques
- Video Surveillance and Tracking Methods
- Magnetic Properties and Applications
- High voltage insulation and dielectric phenomena
- Music and Audio Processing
- Topic Modeling
- Currency Recognition and Detection
- Fault Detection and Control Systems
- Text and Document Classification Technologies
- Non-Destructive Testing Techniques
- Speech Recognition and Synthesis
- Structural Load-Bearing Analysis
- Electricity Theft Detection Techniques
- Quantum many-body systems
- Electrical and Bioimpedance Tomography
- Natural Language Processing Techniques
- Educational Robotics and Engineering
- Bayesian Modeling and Causal Inference
Huazhong University of Science and Technology
2023-2025
Southwest University
2014-2024
Fuzhou University
2022-2024
Xidian University
2024
National University of Defense Technology
2024
Shanghai Electric (China)
2024
Inner Mongolia Electric Power (China)
2024
Hubei University of Education
2020-2023
University of Michigan
2019-2023
Kuaishou (China)
2023
As the search continues for useful applications of noisy intermediate scale quantum devices, variational simulations fermionic systems remain one most promising directions. Here, we perform a series chemistry largest which involved dozen qubits, 78 two-qubit gates, and 114 one-qubit gates. We model binding energy ${\rm H}_6$, H}_8$, H}_{10}$ H}_{12}$ chains as well isomerization diazene. also demonstrate error-mitigation strategies based on $N$-representability dramatically improve effective...
Realizing the potential of quantum computing requires sufficiently low logical error rates1. Many applications call for rates as 10-15 (refs. 2-9), but state-of-the-art platforms typically have physical near 10-3 10-14). Quantum correction15-17 promises to bridge this divide by distributing information across many qubits in such a way that errors can be detected and corrected. Errors on encoded qubit state exponentially suppressed number grows, provided are below certain threshold stable...
Several research studies have been conducted on multi-label classification algorithms for text and images, but few users. Moreover, the existing user algorithm does not provide an effective representation of users, it is difficult to use directly in social media scenarios. By analyzing complex networks, this paper aims achieve users based single-label classification. Considering limitations research, proposes a topic method heterogeneous networks as well community detection. The model...
Pu-erh tea is a healthy beverage rich in phytochemicals, and its effect on the risk of inducing circadian rhythm disorders (CRD) unclear. In this study, mice were given water or 0.25% (w/v) for 7 weeks, followed by 40 day disruption light/dark cycle. CRD caused dysregulation neurotransmitter secretion clock gene oscillations, intestinal inflammation, microbes metabolites. boosted indole 5-hydroxytryptamine pathways tryptophan metabolism via gut-liver-brain axis. Furthermore, metabolites...
Acoustic sensing is one of the most widely used modalities to intelligently assess environment. In particular, ultra-low power (ULP) always-on voice activity detection (VAD) gaining attention as an enabling technology for IoT platforms. many practical applications, acoustic events-of-interest occur infrequently. Therefore, system consumption typically dominated by wakeup detector, while remainder power-gated vast majority time. A previous detector [1] consumed just 12nW but could not process...
This article presents a voice and acoustic activity detector that uses mixer-based architecture ultra-low-power neural network (NN)-based classifier. By sequentially scanning 4 kHz of frequency bands down-converting to below 500 Hz, feature extraction power consumption is reduced by 4×. The NN processor employs computational sprinting, enabling 12× reduction. system also features inaudible signature detection for intentional remote silent wakeup the while re-using subset same components....
Glucolipid metabolism, nitrogen and inflammation are closely related to circadian rhythm disorder (CRD). Ripened Pu-erh tea (RPT) shows significant antidyslipidemic, antihyperurecemic, anti-inflammatory effects. However, it is unclear whether healthy population affected by CRD long-term consumption of RPT can alleviate it. To investigate this problem, mice were pretreated with (0.25%, w/v) for 60 days then subjected 40 days. Our results indicated that showed obesity, the intestinal liver...
This case study examines the design and practice of Problem-Based Learning (PBL) in health communication teaching at universities, using Peking University's "Youth Talk" tobacco control campaign as an exemplar. The systematically investigates implementation process effectiveness evaluation through SWOT analysis, questionnaire surveys, media monitoring. Findings reveal that PBL significantly enhances students' cognition attitudes, effectively cultivates practical capabilities, promotes youth...
Heart failure hospitalization is a severe burden on healthcare. How to predict and therefore prevent readmission has been significant challenge in outcomes research. To address this, we propose deep learning approach from clinical notes. Unlike conventional methods that use structured data for prediction, leverage the unstructured notes train models based convolutional neural networks (CNN). We then trained classify potentially high-risk admissions/patients. For evaluation, CNNs using...
As an important part of the ubiquitous power Internet Things, distribution Things can further improve automation and informatization level network. The reliability measurement data low-voltage terminal unit, as sensing unit layer has a great impact on fault processing advanced applications Things. self-check equipment working status monitoring main station struggle to identify abnormality data. Aiming at this problem, abnormal detection identification recognition method is proposed basis...
Processing self-related material recruits similar neural networks regardless of whether the self-relevance is made explicit or not. However, when considering mechanisms that distinctly underlie cognitive and affective components self-reflection, it still unclear same are involved self-reflection implicit, how these may be modulated by individual personality traits, such as self-esteem. In present functional MRI study, 25 participants were exposed to positive negative words varied with...
Naive Bayes (NB) is an extremely simple and remarkably effective approach to classification learning, but its conditional independence assumption rarely holds true in real-world applications. Attribute weighting known as a flexible model via assigning each attribute different weight discriminatively improve NB. approaches can fall into two broad categories: filters wrappers. Wrappers receive bigger boost terms of accuracy compared with filters, the time complexity wrappers much higher than...
Learning-based video compression has been extensively studied over the past years, but it still limitations in adapting to various motion patterns and entropy models. In this paper, we propose multi-mode (MMVC), a block wise mode ensemble deep framework that selects optimal for feature domain prediction different patterns. Proposed multi-modes include ConvLSTM-based prediction, optical flow conditioned propagation address wide range of cases from static scenes without apparent motions...
The ultra-high-frequency (UHF) method is used to analyze the insulation condition of electric equipment by detecting UHF electromagnetic (EM) waves excited partial discharge (PD). As part detection system, sensor determines system performance in signal extraction and recognition. In this paper, a antenna with fractal structure for PD switchgears was designed means modeling, simulation optimization. This sensor, flat-plate structure, had two resonance frequencies 583 MHz 732 MHz. laboratory,...