检索结果-南通市图书馆

IEEE/CAA Journal of Automatica Sinica 2024年第3期11卷 661-672页

作者： Dan Zhang Jiabin Hu Jun Cheng Zheng-Guang Wu Huaicheng Yan IEEE zhejiang university of technology Guangxi Normal university zhejiang university the East China university of science and technology

This paper proposes a new global fixed-time sliding mode control strategy for the trajectory tracking control of uncertain robotic ***,a fixed-time disturbance observer(FTDO) is designed to deal with the adverse effects of model uncertainties and external disturbances in the manipulator *** an adaptive scheme is used and the adaptive FTDO(AFTDO) is developed,so that the priori knowledge of the lumped disturbance is not ***,a new non-singular fast terminal sliding mode(NFTSM) surface is designed by using an arctan function,which helps to overcome the singularity problem and enhance the robustness of the *** on the estimation of the lumped disturbance by the AFTDO,a fixed-time non-singular fast terminal sliding mode controller(FTNFTSMC)is developed to guarantee the trajectory tracking errors converge to zero within a fixed *** settling time is independent of the initial state of the *** addition,the stability of the AFTDO and FTNFTSMC is strictly proved by using Lyapunov ***,the fixed-time NFESM(FTNFTSM) algorithm is validated on a 2-link manipulator and comparisons with other existing sliding mode controllers(SMCs) are *** comparative results confirm that the FTNFTSMC has superior control performance.

关键词： Disturbance observer(DO) fixed-time non-singular sliding mode control robotic manipulator trajectory tracking

Large language model and domain-specific model collaboration for smart education

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information technology & Electronic Engineering 2024年第3期25卷 333-341页

作者： Yawei LUO Yi YANG School of Software technology Zhejiang UniversityNingbo 315048China College of Computer science and technology Zhejiang UniversityHangzhou 310027China

In this paper,we introduce the large language model and domain-specific model collaboration(LDMC)framework designed to enhance smart *** LDMC framework leverages the comprehensive and versatile knowledge of large domain-general models,combines it with the specialized and disciplinary knowledge from small domainspecific models(DSMs),and incorporates pedagogy knowledge from learning theory *** integration yields multiple knowledge representations,fostering personalized and adaptive educational *** explore various applications of the LDMC framework in the context of smart education.

关键词： smart language knowledge

Molecular Chaperone-Dependent Polymer Translocation through Nanopores: The Effects of Chaperone Concentration and Chaperone-Polymer Interaction

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Polymer science 2024年第1期42卷 125-132,I0011页

作者： Chang-Sheng Zuo Kang Wang Li-Zhen Sun Ting-Ting Sun Department of Physics Zhejiang University of Science and TechnologyHangzhou 310023China Department of Applied Physics Zhejiang University of TechnologyHangzhou 310014China

The polymer translocation through a nanopore from a donor space(or named cis side) to a receiver space(trans side) in the chaperone-induced crowded environment has attracted increasing attention in recent years due to its significance in biological systems and technological applications. In this work, we mainly focus on the effects of chaperone concentration and chaperone-polymer interaction on the polymer translocation. By assuming the polymer translocation to be a quasi-equilibrium process, the free energy F of the polymer can be estimated by Rosenbluth-Rosenbluth method and then the translocation time τ can be calculated by Fokker-Plank equation based on the obtained free energy landscape. Our calculation results show that the translocation time can be controlled by independently tuning the chaperone concentration and chaperone-polymer interaction at the cis side or the trans side. There exists a critical chaperone-polymer attraction ε~*=-0.2 at which the volume exclusion and interaction effects of the chaperone can balance each other. Additionally, we also find that at large chaperone-polymer attraction, the translocation time is mainly governed by the diffusion coefficient of the polymer.

关键词： Polymer translocation Rosenbluth-Rosenbluth method Fokker-Plank equation Free energy landscapes

Multi-exit self-distillation with appropriate teachers

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information technology & Electronic Engineering 2024年第4期25卷 585-599页

作者： Wujie SUN Defang CHEN Can WANG Deshi YE Yan FENG Chun CHEN College of Computer science and technology Zhejiang UniversityHangzhou 310000China

Multi-exit architecture allows early-stop inference to reduce computational cost,which can be used in resource-constrained *** works combine the multi-exit architecture with self-distillation to simultaneously achieve high efficiency and decent performance at different network ***,existing methods mainly transfer knowledge from deep exits or a single ensemble to guide all exits,without considering that inappropriate learning gaps between students and teachers may degrade the model performance,especially in shallow *** address this issue,we propose Multi-exit self-distillation with Appropriate TEachers(MATE)to provide diverse and appropriate teacher knowledge for each *** MATE,multiple ensemble teachers are obtained from all exits with different trainable *** exit subsequently receives knowledge from all teachers,while focusing mainly on its primary teacher to keep an appropriate gap for efficient knowledge *** this way,MATE achieves diversity in knowledge distillation while ensuring learning *** results on CIFAR-100,TinyImageNet,and three fine-grained datasets demonstrate that MATE consistently outperforms state-of-the-art multi-exit self-distillation methods with various network architectures.

关键词： Multi-exit architecture Knowledge distillation Learning gap

Structure and dynamical properties during solidification of liquid aluminum induced by cooling and compression

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Physics B 2024年第7期33卷 476-481页

作者：吴旻杨永琪王垚 College of Materials science and Engineering Zhejiang University of TechnologyHangzhou 310014China Shaoxing Research Institute Zhejiang University of TechnologyShaoxing 312000China

The structural transformation from a liquid into a crystalline solid is an important subject in condensed matter physics and materials science. In the present study, first-principles molecular dynamics calculations are performed to investigate the structure and properties of aluminum during the solidification which is induced by cooling and compression. In the cooling process and compression process, it is found that the icosahedral short-range order is initially enhanced and then begin to decay, the face-centered cubic short-range order eventually becomes dominant before it transforms into a crystalline solid.

关键词： first-principles method molecular dynamics short-range order liquid aluminum

同方期刊数据库

Industry-Academic-Education Integration: Innovation and Practice of Training Mode of Applied Statistics Talents in Big Data Era——Taking zhejiang university of science and technology as an Example

在线全文

学校读者我要写书评

暂无评论

Industry-Academic-Education Integration: Innovation and Prac...

A Robust Framework for Multimodal Sentiment Analysis with Noisy Labels Generated from Distributed Data Annotation

作者： Songyan Zhang Lichao Tao Taotao Zheng School of Economics and Management Zhejiang University of Science and Technology zhejiang College of Construction

In the era of big data, new requirements are put forward for Applied Statistical ***-academic-education integration is an effective way to cultivate applied *** topic aims at the existing problems, combined with the practice of zhejiang university of science and technology, carried out the innovation and practice of applied statistics talent training mode.

关键词： Industry-Academic-Education Integration Applied Statistics Talents Training Mode

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & sciences 2024年第6期139卷 2965-2984页

作者： Kai Jiang Bin Cao Jing Fan School of Computer science and technology Zhejiang University of TechnologyHangzhou310023China

Multimodal sentiment analysis utilizes multimodal data such as text,facial expressions and voice to detect people’s *** the advent of distributed data collection and annotation,we can easily obtain and share such multimodal ***,due to professional discrepancies among annotators and lax quality control,noisy labels might be *** research suggests that deep neural networks(DNNs)will overfit noisy labels,leading to the poor performance of the *** address this challenging problem,we present a Multimodal Robust Meta Learning framework(MRML)for multimodal sentiment analysis to resist noisy labels and correlate distinct modalities ***,we propose a two-layer fusion net to deeply fuse different modalities and improve the quality of the multimodal data features for label correction and network ***,a multiple meta-learner(label corrector)strategy is proposed to enhance the label correction approach and prevent models from overfitting to noisy *** conducted experiments on three popular multimodal datasets to verify the superiority of ourmethod by comparing it with four baselines.

关键词： Distributed data collection multimodal sentiment analysis meta learning learn with noisy labels

PURP: A Scalable System for Predicting Short-Term Urban TrafficFlow Based on License Plate Recognition Data

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Big Data Mining and Analytics 2024年第1期7卷 171-187页

作者： Shan Zhang Qinkai Jiang Hao Li Bin Cao Jing Fan School of Computer science and technology Zhejiang University of TechnologyHangzhou 310000China

Accurate and efficient urban traffic flow prediction can help drivers identify road traffic conditions in real-time,consequently helping them avoid congestion and accidents to a certain ***,the existing methods for real-time urban traffic flow prediction focus on improving the model prediction accuracy or efficiency while ignoring the training efficiency,which results in a prediction system that lacks the scalability to integrate real-time traffic flow into the training *** conduct accurate and real-time urban traffic flow prediction while considering the latest historical data and avoiding time-consuming online retraining,herein,we propose a scalable system for Predicting short-term URban traffic flow in real-time based on license Plate recognition data(PURP).First,to ensure prediction accuracy,PURP constructs the spatio-temporal contexts of traffic flow prediction from License Plate Recognition(LPR)data as effective ***,to utilize the recent data without retraining the model online,PURP uses the nonparametric method k-Nearest Neighbor(namely KNN)as the prediction framework because the KNN can efficiently identify the top-k most similar spatio-temporal contexts and make predictions based on these contexts without time-consuming model retraining *** experimental results show that PURP retains strong prediction efficiency as the prediction period increases.

关键词： traffic flow prediction k-Nearest Neighbor(KNN) License Plate Recognition(LPR)data spatio-temporalcontext

Electron-microscopical visualization on the interfacial and crystallographic structures of lithium metal anode

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Chemical Letters 2024年第6期35卷 26-37页

作者： Jiale Zheng Mei Chen Huadong Yuan Jianmin Luo Yao Wang Jianwei Nai Xinyong Tao Yujing Liu College of Materials science and Engineering Zhejiang University of TechnologyHangzhou 310014China College of Mechanical and Energy Engineering Zhejiang University of Science and TechnologyHangzhou 310014China

Lithium(Li)metal anodes(LMAs)have garnered significant attention as a potential solution for developing high-energy density batteries,given their theoretical specific capacity and redox ***,safety concerns and internal cycling stability issues originated from uncontrollable Li dendrite growth have impeded the practical application of *** the interface between Li metal and electrolyte is a crucial process that offers valuable insights into the characteristics and regularity of primary circular *** illustrate the intrinsic characteristic of Li metal batteries(LMBs)in spatial and temporal,it is imperative to employ electron microscopes to characterize the structural components distribution of Li with atomic *** paper summarizes the progress in the characterization and analysis of the interfaces in LMBs with electron microscopes based on the principles of electron-matter ***,future trends and the potential of electron microscopes are also discussed to advance our understanding of LMBs.

关键词： Li metal anode Electron microscopy Visualization High-resolution Battery

State-sensitive convolutional sparse coding for potential biomarker identification in brain signals

在线全文

学校读者我要写书评

暂无评论

science China(Information sciences) 2024年第5期67卷 270-286页

作者： Puli WANG Yu QI Gang PAN College of Computer science and technology Zhejiang University State Key Laboratory of Brain-Machine Intelligence Zhejiang University MOE Frontier science Center for Brain science and Brain-machine Integration Zhejiang University Affiliated Mental Health Center & Hangzhou Seventh People's Hospital Zhejiang University School of Medicine

The identification of prototypical waveforms, such as sleep spindles and epileptic spikes, is crucial for the diagnosis of neurological disorders. These prototypical waveforms are usually recurrently presented in certain brain states, serving as potential biomarkers for clinical evaluations. Convolutional sparse coding(CSC) approaches have demonstrated strength in identifying recurrent patterns in time-series. However,existing CSC approaches do not explicitly explore state-specific patterns, making it difficult to identify state-related biomarkers. To address this problem, we propose state-sensitive CSC to learn state-specific prototypical waveforms. Specifically, we model signals of a certain state with specific waveforms that only appear frequently in this state and background waveforms that are independent of states. Based on this,state-sensitive CSC separates state-specific waveforms from background ones explicitly by incorporating incoherence constraints into optimizations. Experiments with epilepsy brain signals demonstrate that our approach can effectively identify prototypical waveforms in pre-ictal states, providing potential biomarkers for seizure prediction. Our approach provides a promising tool for automatic biomarker candidate identification.

关键词： convolutional sparse coding prototypical waveforms state-specific dictionary biomarker identification neural signal processing