检索结果-南通市图书馆

维普期刊数据库

FSCIL-EACA: Few-Shot Class-Incremental Learning Network Based on Embedding Augmentation and Classifier adaptation for Image Classification

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2024年第1期33卷 139-152页

作者： Ruru ZHANG Haihong E Meina SONG School of Computer Science Beijing University of Posts and Telecommunications Education Department Information Network Engineering Research Center Beijing University of Posts and Telecommunications

The ability to learn incrementally is critical to the long-term operation of AI systems. Benefiting from the power of few-shot class-incremental learning(FSCIL), deep learning models can continuously recognize new classes with only a few samples. The difficulty is that limited instances of new classes will lead to overfitting and exacerbate the catastrophic forgetting of the old classes. Most previous works alleviate the above problems by imposing strong constraints on the model structure or parameters, but ignoring embedding network transferability and classifier adaptation(CA), failing to guarantee the efficient utilization of visual features and establishing relationships between old and new classes. In this paper, we propose a simple and novel approach from two perspectives: embedding bias and classifier bias. The method learns an embedding augmented(EA) network with cross-class transfer and class-specific discriminative abilities based on self-supervised learning and modulated attention to alleviate embedding bias. Based on the adaptive incremental classifier learning scheme to realize incremental learning capability,guiding the adaptive update of prototypes and feature embeddings to alleviate classifier bias. We conduct extensive experiments on two popular natural image datasets and two medical datasets. The experiments show that our method is significantly better than the baseline and achieves state-of-the-art results.

关键词： Training adaptation models Visualization Image recognition Prototypes Self-supervised learning Power capacitors

同方期刊数据库

Research on Modeling and Parameter Identification of Comprehensive Load Model of Distribution Network in Industrial Park

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2024年第1期33卷 264-273页

作者： Tingling WANG Xiaohe YAN School of Electric Engineering North China University of Water Resources and Electric Power Wenzhou Polytechnic

With the rapid development of industrial parks, its load model research has become a hot spot. In order to study the load of power system in industrial park, based on the characteristics of the industrial park load, a comprehensive load admittance static model with full voltage range adaptability is considered, and a comprehensive load model of distribution network of the industrial park is established. A complete parameter identification of the model is carried out through chaos particle swarm optimization. The simulation results show that the model can effectively describe the load characteristics of the distribution network in industrial parks.

关键词： Chaos adaptation models Analytical models Parameter estimation Induction motors Simulation Distribution networks

同方期刊数据库

BAD-FM:Backdoor Attacks Against Factorization-Machine Based Neural Network for Tabular Data Prediction

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2024年第4期33卷 1077-1092页

作者： Lingshuo MENG Xueluan GONG Yanjiao CHEN College of Electrical Engineering Zhejiang University School of Computer Science Wuhan University

Backdoor attacks pose great threats to deep neural network models. All existing backdoor attacks are designed for unstructured data(image, voice, and text), but not structured tabular data, which has wide real-world applications, e.g., recommendation systems, fraud detection, and click-through rate prediction. To bridge this research gap, we make the first attempt to design a backdoor attack framework, named BAD-FM, for tabular data prediction models. Unlike images or voice samples composed of homogeneous pixels or signals with continuous values, tabular data samples contain well-defined heterogeneous fields that are usually sparse and discrete. Tabular data prediction models do not solely rely on deep networks but combine shallow components(e.g., factorization machine, FM) with deep components to capture sophisticated feature interactions among fields. To tailor the backdoor attack framework to tabular data models, we carefully design field selection and trigger formation algorithms to intensify the influence of the trigger on the backdoored model. We evaluate BAD-FM with extensive experiments on four datasets, i.e.,HUAWEI, Criteo, Avazu, and KDD. The results show that BAD-FM can achieve an attack success rate as high as 100%at a poisoning ratio of 0.001%, outperforming baselines adapted from existing backdoor attacks against unstructured data models. As tabular data prediction models are widely adopted in finance and commerce, our work may raise alarms on the potential risks of these models and spur future research on defenses.

关键词： adaptation models Systematics Frequency modulation Finance Predictive models Prediction algorithms Data models

Domain Adaptive Learning with Multi-Granularity Features for Unsupervised Person Re-identification

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2022年第1期31卷 116-128页

作者： FU Lihua DU Yubin DING Yu WANG Dan JIANG Hanxu ZHANG Haitao Beijing University of Technology

Unsupervised person re-identification(Re-ID) aims to improve the model's scalability and obtain better Re-ID results in the unlabeled data domain. In this paper, we propose an unsupervised person Re-ID method based on multi-granularity feature representation and domain adaptive learning, which can effectively improve the performance of unsupervised person re-identification. The multi-granularity feature extraction module integrates global and local information of different granularity to obtain the multi-granularity person feature representation with rich discriminative information. The source domain classification module learns the labeled source dataset classification and obtains the person's discriminative knowledge in the source domain. On this basis, the domain adaptive module further considers the difference between the target domain and the source domain to learn adaptively for the model. Experiments on multiple public datasets show that the proposed method can achieve a competitive performance among other stateof-the-art unsupervised Re-ID methods.

关键词： Feature extraction adaptation models Adaptive learning Transfer learning Semantics Training Scalability

Self-Adaptive Discrete Cuckoo Search Algorithm for the Service Routing Problem with Time Windows and Stochastic Service Time

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2023年第4期32卷 920-931页

作者： ZHANG Guoyun WU Meng LI Wujing OU Xianfeng XIE Wenwu School of Information Science and Engineering Hunan Institute of Science and Technology

Making house calls is very crucial to deal with the competitive pressures of the service business and to improve service quality. We design a model called service routing problem with time windows and stochastic service time(SRPTW-SST) that is based on vehicle routing problem with time windows. A self-adaptive discrete cuckoo search algorithm with genetic mechanism(sDCSGM) is proposed for the model SRPTW-SST. Moreover,we design a selection mechanism to improve the logicality of the algorithm based on the strong randomness of the Lévy flight. We introduce a genetic mechanism and design a neighborhood search mechanism for improving the robustness of the algorithm. In addition, an adaptive parameter adjustment method is designed to eliminate the impact of fixed parameters. The experimental results show that the sDCS-GM algorithm is more robust and effective than the state-of-the-art methods.

关键词： adaptation models Vehicle routing Stochastic processes Routing Search problems Genetics Robustness

Cross Modal Adaptive Few-Shot Learning Based on Task Dependence

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2023年第1期32卷 85-96页

作者： DAI Leichao FENG Lin SHANG Xinglin SU Han School of Computer Science Sichuan Normal University

Few-shot learning(FSL) is a new machine learning method that applies the prior knowledge from some different domains tasks. The existing FSL models of metric-based learning have some drawbacks,such as the extracted features cannot reflect the true data distribution and the generalization ability is weak. In order to solve the problem in the present, we developed a model named cross modal adaptive few-shot learning based on task dependence(COOPERATE for short). A feature extraction and task representation method based on task condition network and auxiliary co-training is proposed. Semantic representation is added to each task by combining both visual and textual features. The measurement scale is adjusted to change the property of parameter update of the algorithm. The experimental results show that the COOPERATE has the better performance comparing with all approaches of the monomode and modal alignment FSL.

关键词： Measurement adaptation models Visualization Semantics Machine learning Feature extraction Data models

Adaptive Tensor Rank Approximation for Multi-View Subspace Clustering

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2023年第4期32卷 840-853页

作者： SUN Xiaoli HAI Yang ZHANG Xiujun XU Chen College of Mathematics and Statistics Shenzhen University School of Electronic and Communication Engineering Shenzhen Polytechnic National Center for Applied Mathematics Shenzhen (NCAMS)

Multi-view subspace clustering under a tensor framework remains a challenging problem, which can be potentially applied to image classification, impainting, denoising, etc. There are some existing tensor-based multi-view subspace clustering models mainly making use of the consistency in different views through tensor nuclear norm(TNN). The diversity which means the intrinsic difference in individual view is always ignored. In this paper, a new tensorial multi-view subspace clustering model is proposed, which jointly exploits both the consistency and diversity in each view. The view representation is decomposed into view-consistent part(low-rank part) and view-specific part(diverse part). A tensor adaptive logdeterminant regularization(TALR) is imposed on the low-rank part to better relax the tensor multi-rank, and a view-specific sparsity regularization is applied on the diverse part to ensure connectedness property. Although the TALR minimization is not convex, it has a closedform analytical solution and its convergency is validated mathematically. Extensive evaluations on six widely used clustering datasets are executed and our model is demonstrated to have the superior performance.

关键词： adaptation models Tensors Closed-form solutions Noise reduction Clustering algorithms Minimization Mathematical models

Teacher-Student Training Approach Using an Adaptive Gain Mask for LSTM-Based Speech Enhancement in the Airborne Noise Environment

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2023年第4期32卷 882-895页

作者： HUANG Ping WU Yafeng School of Power and Energy Northwestern Polytechnical University

Research on speech enhancement algorithms in the airborne environment is of great significance to the security of airborne systems. Recently, the research focus of speech enhancement has turned from conventional unsupervised algorithms, like the log minimum mean square error estimator(log-MMSE), to the state-ofthe-art masking-based long short-term memory(LSTM)method. However, each method has its characteristics and limitations, so they cannot always handle noise well. Besides, the requirements of clean speech and noise data for training a supervised speech enhancement model are difficult to satisfy in the real-world airborne ***, in this work, to fully utilize the advantages of those two different methods without any data restrictions,we propose a novel adaptive gain mask(AGM) based teacher-student training approach for speech enhancement. In our method, the AGM, as a robust learning target for the student model, is devised by incorporating the estimated ideal ratio mask from the teacher model into the procedure of the log-MMSE approach. To get an appropriate tradeoff between the two methods, we adaptively update the AGM using a recursive weighting coefficient. Experiments on the real airborne data show that the proposed AGM-based method outperforms other baselines in terms of all essential objective metrics evaluated in this paper.

关键词： Training Measurement Deep learning adaptation models Atmospheric modeling Mean square error methods Interference

Improved Cross-Corpus Speech Emotion Recognition Using Deep Local Domain adaptation

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2023年第3期32卷 640-646页

作者： ZHAO Huijuan YE Ning WANG Ruchuan College of Internet of Things Nanjing University of Posts and Telecommunications College of Computer and Software Nanjing Vocational University of Industry Technology College of Computer Nanjing University of Posts and Telecommunications Jiangsu High Technology Research Key Laboratory for Wireless Sensor Network

Due to the scarcity of high-quality labeled speech emotion data, it is natural to apply transfer learning to emotion recognition. However, transfer learning-based speech emotion recognition becomes more challenging because of the complexity and ambiguity of emotion. Domain adaptation based on maximum mean discrepancy considers marginal alignment of source domain and target domain, but not pay regard to class prior distribution in both domains, which results in the reduction of transfer efficiency. In order to address the problem, this study proposes a novel cross-corpus speech emotion recognition framework based on local domain adaption. A category-grained discrepancy is used to evaluate the distance between two relevant domains. According to research findings, the generalization ability of the model is enhanced by using the local adaptive *** with global adaptive and non-adaptive methods, the effectiveness of cross-corpus speech emotion recognition is significantly improved.

关键词： Emotion recognition adaptation models Transfer learning Speech recognition Complexity theory