检索结果-南通市图书馆

Pushing one pair of labels apart each time in multi-label learning: from single positive to full labels

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2025年

作者： Xiang LI Xinrui WANG Songcan CHEN MIIT Key laboratory of pattern Analysis and Machine Intelligence College of Computer Science and Technology/College of Artificial Intelligence Nanjing University of Aeronautics and Astronautics

In multi-label learning (MLL), it is extremely challenging to accurately annotate every appearing object due to expensive costs and limited knowledge. When facing such a challenge, a more practical and cheaper alternative should be single positive multi-label learning (SPMLL), where only one positive label needs to be provided per sample. Existing SPMLL methods usually assume unknown labels as negatives, which inevitably introduces false negatives as noisy labels. More seriously, binary cross entropy (BCE) loss is often used for training, which is notoriously not robust to noisy labels. To mitigate this issue, we customize an objective function for SPMLL by pushing only one pair of labels apart each time to suppress the domination of negative labels, which is the main culprit of fitting noisy labels in SPMLL. To further combat such noisy labels, we explore the high-rankness of the label matrix, which can also push apart different labels. By directly extending from SPMLL to MLL with full labels, a unified loss applicable to both settings is derived. As a byproduct, the proposed loss can alleviate the imbalance inherent in MLL. Experiments on real datasets demonstrate that the proposed loss not only performs more robustly to noisy labels for SPMLL but also works well for full labels. Besides, we empirically discover that high-rankness can mitigate the dramatic performance drop in SPMLL. Most surprisingly, even without any regularization or fine-tuned label correction, only adopting our loss defeats state-of-the-art SPMLL methods on CUB, a dataset that severely lacks labels.

关键词：

Degeneracy from Twisted Cubic Under Two Views

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2010年第5期25卷 916-924页

作者：吴毅红兰添胡占义 national laboratory of pattern recognition Institute of AutomationChinese Academy of Sciences

Fundamental matrix,drawing geometric relationship between two images,plays an important role in 3- dimensional computer *** configurations of the space points and the two camera optical centers affect stability of computation for fundamental *** order to robustly estimate fundamental matrix,it is necessary to study these degenerate *** analyze all the possible degenerate configurations caused by twisted cubic and give the corresponding degenerate rank for each *** with general degeneracies,the previous ruled quadric degeneracy and the homography degeneracy,are also reported.

关键词： fundamental matrix twisted cubic ruled quadric degenerate configuration

Linguistic Theory Based Contextual Evidence Mining for Statistical Chinese Co-Reference Resolution

在线全文

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2007年第4期22卷 608-617页

作者：赵军刘非凡 national laboratory of pattern recognition Institute of AutomationChinese Academy of Sciences

Under statistical learning framework, the paper focuses on how to use traditional linguistic findings on anaphora resolution as a guide for mining and organizing contextual features for Chinese co-reference resolution. The main achievements are as follows. （1） In order to simulate ＂syntactic and semantic parallelism factor＂, we extract ＂bags of word form and POS＂ feature and ＂bag of seines＂ feature from the contexts of the entity mentions and incorporate them into the baseline feature set. （2） Because it is too coarse to use the feature of bags of word form, POS tag and seme to determine the syntactic and semantic parallelism between two entity mentions, we propose a method for contextual feature reconstruction based on semantic similarity computation, in order that the reconstructed contextual features could better approximate the anaphora resolution factor of ＂Syntactic and Semantic Parallelism Preferences＂. （3） We use an entity-mention-based contextual feature representation instead of isolated word-based contextual feature representation, and expand the size of the contextual windows in addition, in order to approximately simulate ＂the selectional restriction factor＂ for anaphora resolution. The experiments show that the multi-level contextual features are useful for co-reference resolution, and the statistical system incorporated with these features performs well on the standard ACE datasets.

关键词： natural language processing information extraction co-reference resolution anaphora resolution

Reconstructing forest and grassland cover changes in China over the past millennium

在线全文

学校读者我要写书评

暂无评论

Science China Earth Sciences 2025年第1期68卷 94-110页

作者： Fanneng HE Fan YANG Yafei WANG Key laboratory of Land Surface pattern and Simulation Institute of Geographic Sciences and Natural Resources ResearchChinese Academy of SciencesBeijing100101China Key Research Institute of Yellow River Civilization and Sustainable Development Henan UniversityKaifeng475001China University of Chinese Academy of Sciences Beijing100049China

Historical forest and grassland cover changes not only are critical indicators for quantifying ecological and environmental change processes but also serve as fundamental data for long-term climate change simulations and terrestrial ecosystem carbon emission ***,because of limitations in historical data,quantitative estimations and spatially gridded reconstructions of these changes remain challenging,necessitating further methodological *** study focused on China's present-day land area over the past millennium,objectively capturing the characteristics and drivers of forest and grassland cover *** this basis,using the forest transition theory and the space-for-time substitution method,we depicted the historical deforestation process as an inverted“S”curve and developed a model to reconstruct historical forest area changes based on the functional relationship between the forest area and population size ***,a gridded forest allocation model was established on the basis of deforestation *** the grassland cover,we implemented regionspecific methods,such as the cropland area deduction method and the habitat constraint method,to quantitatively reconstruct historical ***,we obtained provincial forest and grassland area changes over the past millennium and mapped 10-km-resolution gridded data of forest and grassland *** results indicated the following.(1)The methods developed using population data as a proxy objectively reproduced the spatiotemporal evolution of forest and grassland cover in China over the past *** feasible methods offer a novel pathway for the quantitative reconstruction of historical forest and grassland cover changes.(2)The data indicated that China's forest area generally decreased over the past millennium,characterized by a“decrease-then-increase”*** forest area experienced three distinct phases:a slow decline(AD 1000–1650),a rapid decline(AD 16

关键词： Land use and land cover change Forest Grassland Quantitative reconstruction Gridded reconstruction China historical period

Systematic analysis of innate immune-related genes in the silkworm:Application to antiviral research

在线全文

学校读者我要写书评

暂无评论

Insect Science 2025年第1期32卷 151-171页

作者： Junming Xia Ruoxuan Peng Shigang Fei Mian Muhammad Awais Wenxuan Lai Yigui Huang Hailin Wu Yue Yu Lingying Liang Luc Swevers Jingchen Sun Min Feng Guangdong Provincial Key laboratory of Agro-Animal Genomics and Molecular Breeding College of Animal ScienceSouth China Agricultural UniversityGuangzhouChina Insect Molecular Genetics and Biotechnology National Center for Scientific Research DemokritosInstitute of Biosciences and ApplicationsAthensGreece

The silkworm,a crucial model organism of the Lepidoptera,offers an excellent platform for investigating the molecular mechanisms underlying the innate immune response of insects toward *** the years,researchers worldwide have identified numerous immune-related genes in ***,these identified silkworm immune genes are not well classified and not well known to the scientific *** the availability of the latest genome data of silkworms and the extensive research on silkworm immunity,it has become imperative to systematically categorize the immune genes of silkworms with different database *** this study,we present a meticulous organization of prevalent immune-related genes in the domestic silkworm,using the SilkDB 3.0 database as a reliable source for updated gene ***,utilizing the available data,we classify the collected immune genes into distinct categories:pattern recognition receptors,classical immune pathways,effector genes and ***-depth data analysis has enabled us to predict some potential antiviral ***,we performed antiviral experiments on selected genes,exploring their impact on Bombyx mori nucleopolyhedrovirus *** outcomes of this research furnish novel insights into the immune genes of the silkworm,consequently fostering advancements in the field of silkworm immunity research by establishing a comprehensive classification and functional understanding of immune-related genes in the *** study contributes to the broader understanding of insect immune responses and opens up new avenues for future investigations in the domain of host–pathogen interactions.

关键词： antiviral defense BmNPV Bombyx mori innate immunity SilkDB 3.0

Statistical learning based facial animation

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

Journal of Zhejiang University-Science C(Computers and Electronics) 2013年第7期14卷 542-550页

作者： Shibiao XU Guanghui MA Weiliang MENG Xiaopeng ZHANG national laboratory of pattern recognition Institute of Automation Chinese Academy of Sciences

To synthesize real-time and realistic facial animation, we present an effective algorithm which combines image- and geometry-based methods for facial animation simulation. Considering the numerous motion units in the expression coding system, we present a novel simplified motion unit based on the basic facial expression, and construct the corresponding basic action for a head model. As image features are difficult to obtain using the performance driven method, we develop an automatic image feature recognition method based on statistical learning, and an expression image semi-automatic labeling method with rotation invariant face detection, which can improve the accuracy and efficiency of expression feature identification and training. After facial animation redirection, each basic action weight needs to be computed and mapped automatically. We apply the blend shape method to construct and train the corresponding expression database according to each basic action, and adopt the least squares method to compute the corresponding control parameters for facial animation. Moreover, there is a pre-integration of diffuse light distribution and specular light distribution based on the physical method, to improve the plausibility and efficiency of facial rendering. Our work provides a simplification of the facial motion unit, an optimization of the statistical training process and recognition process for facial animation, solves the expression parameters, and simulates the subsurface scattering effect in real time. Experimental results indicate that our method is effective and efficient, and suitable for computer animation and interactive applications.

关键词： Facial animation Motion unit Statistical learning Realistic rendering Pre-integration

Multi-wavelength plasmonic optoelectronic memristor for reconfigurable logic operations and mixed-color pattern recognition

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2025年第2期68卷 5-14页

作者： Jiaqi HAN Zhuangzhuang LI Tao ZENG Xuanyu SHAN Riya SU Ya LIN Zhongqiang WANG Xiaoning ZHAO Shencheng FU Haiyang XU Yichun LIU Key laboratory for UV Light-Emitting Materials and Technology (Northeast Normal University) Ministry of Education Department of Materials Science and Engineering National University of Singapore

Optoelectronic memristors have emerged as promising hardware platforms for highly efficient image perception owing to their low crosstalk, fast response, and low power consumption. However, the development of optoelectronic memristors with multi-wavelength recognition capabilities remains constrained by the limitations of their inherent bandgap. In this work, a multi-wavelength-modulated optoelectronic memristor that operates based on the localized surface plasmon resonance effect is developed using silver（Ag）-titanium dioxide（TiO2）/polymethyl methacrylate（PMMA） nanocomposites. The SET voltage can be gradually reduced by modulating the optical wavelength of visible irradiation. On this basis, reconfigurable IMP and False logic operations have been implemented in the memristive logic circuit. The neuromorphic visual preprocessing and recognition function for mixed-color patterns is demonstrated based on wavelength-dependent resistive switching behaviors. The proposed optoelectronic memristor enables high-density storage and in-memory computing.

关键词： optoelectronic memristor localized surface plasmon resonance multi-wavelength resistive switching color pattern recognition

同方期刊数据库

An enhancer–promoter-transcription factor module orchestrates plant immune homeostasis by constraining camalexin biosynthesis

在线全文

学校读者我要写书评

暂无评论

Molecular Plant 2025年第1期18卷 95-113页

作者： Ying Zhang Meng Tang Yi Zhang Qinglin Cheng Lijiang Liu Wei Chen Jiatao Xie Jiasen Cheng Yanping Fu Bo Li Daohong Jiang Xiao Yu national Key laboratory of Agricultural Microbiology Huazhong Agricultural UniversityWuhanHubei 430070China The Provincial Key Lab of Plant Pathology of Hubei Province College of Plant Science and TechnologyHuazhong Agricultural UniversityWuhanHubei 430070China Hubei Hongshan laboratory WuhanHubei 430070China national Key laboratory of Crop Genetic Improvement and national center of Plant Gene Research(Wuhan) Huazhong Agricultural UniversityWuhan 430070China Oil Crops Research Institute of the Chinese Academy of Agricultural Sciences Key Laboratory of Biology and Genetic Improvement of Oil Crops of the Ministry of AgricultureWuhan 430062China

Effective plant defense against pathogens relies on highly coordinated regulation of immune gene ***,as cis-regulatory elements,are indispensable determinants of dynamic gene regulation,but the molecular functions in plant immunity are not well *** this study,we identified a novel enhancer,CORE pattern-INDUCED ENHANCER 35(CPIE35),which is rapidly activated upon pathogenic elicitation and negatively regulates antifungal resistance through modulating WRKY15 *** immune activation,CPIE35 activates the transcription of WRKY15 by forming chromatin loops with the promoter of WRKY15 in a WRKY18/40/60-,WRKY33-,and MYC2-dependent ***15 directly binds to the promoters of PAD3 and GSTU4,suppressing their expression and leading to reduced camalexin synthesis and ***,CPIE35 region is evolutionarily conserved among Brassicaceae plants,and the CPIE35-WRKY15 module exerts similar functions in Brassica napus to negatively regulate antifungal *** work reveals the“enhancer–promoter-transcription factor”regulatory mechanism in maintenance of immune homeostasis,highlighting the importance and conserved role of enhancers in fine-tuning immune gene expression in plants.

关键词： plant innate immunity enhancers antifungal resistance transcriptional regulation cis-regulatory elements transcription factors

VFM： Visual Feedback Model for Robust Object recognition

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

Journal of Computer Science & Technology 2015年第2期30卷 325-339页

作者：王冲黄凯奇 national laboratory of pattern recognition Institute of Automation Chinese Academy of Sciences Beijing 100190 China

Object recognition, which consists of classification and detection, has two important attributes for robustness： 1） closeness： detection windows should be as close to object locations as possible, and 2） adaptiveness： object matching should be adaptive to object variations within an object class. It is difficult to satisfy both attributes using traditional methods which consider classification and detection separately; thus recent studies propose to combine them based on confidence contextualization and foreground modeling. However, these combinations neglect feature saliency and object structure, and biological evidence suggests that the feature saliency and object structure can be important in guiding the recognition from low level to high level. In fact, object recognition originates in the mechanism of ＂what＂ and ＂where＂ pathways in human visual systems. More importantly, these pathways have feedback to each other and exchange useful information, which may improve closeness and adaptiveness. Inspired by the visual feedback, we propose a robust object recognition framework by designing a computational visual feedback model （VFM） between classification and detection. In the ＂what＂ feedback, the feature saliency from classification is exploited to rectify detection windows for better closeness; while in the ＂where＂ feedback, object parts from detection are used to match object structure for better adaptiveness. Experimental results show that the ＂what＂ and ＂where＂ feedback is effective to improve closeness and adaptiveness for object recognition, and encouraging improvements are obtained on the challenging PASCAL VOC 2007 dataset.

关键词： object recognition object classification object detection visual feedback