检索结果-内蒙古大学图书馆

VideoLT: Large-scale Long-tailed Video Recognition

学校读者我要写书评

暂无评论

VideoLT: Large-scale Long-tailed Video Recognition

International Conference on Computer Vision (ICCV)

作者： Xing Zhang Zuxuan Wu Zejia Weng Huazhu Fu Jingjing Chen Yu-Gang Jiang Larry Davis Academy for Engineering and Technology Fudan University Shanghai Key Lab of Intel. Info. Processing School of Computer Science Fudan University Shanghai Collaborative Innovation Center on Intelligent Visual Computing Inception Institute of Artificial Intelligence University of Maryland

ISBN: (纸本)9781665428132

lab.l distributions in real-world are oftentimes long-tailed and imbalanced, resulting in biased models towards dominant lab.ls. While long-tailed recognition has been extensively studied for image classification tasks, limited effort has been made for the video domain. In this paper, we introduce VideoLT, a large-scale long-tailed video recognition dataset, as a step toward real-world video recognition. VideoLT contains 256,218 untrimmed videos, annotated into 1,004 classes with a long-tailed distribution. Through extensive studies, we demonstrate that state-of-the-art methods used for long-tailed image recognition do not perform well in the video domain due to the additional temporal dimension in videos. This motivates us to propose FrameStack, a simple yet effective method for long-tailed video recognition. In particular, FrameStack performs sampling at the frame-level in order to balance class distributions, and the sampling ratio is dynamically determined using knowledge derived from the network during training. Experimental results demonstrate that FrameStack can improve classification performance without sacrificing the overall accuracy. Code and dataset are availab.e at: https://***/17Skye17/VideoLT.

关键词： Training Knowledge engineering Computer vision Adaptation models Image recognition Codes Computational modeling

Learning to Undersampling for Class Imbalanced Credit Risk Forecasting

学校读者我要写书评

暂无评论

Learning to Undersampling for Class Imbalanced Credit Risk F...

IEEE International Conference on Data Mining (ICDM)

作者： Jianfeng Chi Guanxiong Zeng Qiwei Zhong Ting Liang Jinghua Feng Xiang Ao Jiayu Tang Alibaba Group CN Key Lab of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing CN University of Chinese Academy of Sciences Beijing CN

ISBN: (数字)9781728183169

ISBN: (纸本)9781728183176

Credit risk forecasting generally aims to evaluate the default probability of users in financial service. It is usually regarded as a binary classification problem, which suffers from the severe class-imbalance problem due to the extremely limited throngs and the concept drift problem brought by the delayed verification. In this paper, we investigate these problems in credit risk forecasting and propose a semi-supervised meta-learning based approach called TRUST (TRainable Undersampling with Self Training) to resolve. First, it decides whether to sample the data through meta-learning based reinforcement learning. Secondly, it learns the distribution of the data that have not yet shown financial performance via self-training and updates the model trained in the first step. Finally, the updated model is evaluated on the validation dataset, the result of which will be fed back through the evaluator. These three steps will be iterated until the model converges. With the real-world industrial dataset containing 1.75 million users, we investigate the effectiveness of our method. Experimental results exhibit that the proposed method is able to improve AP over 5.94% on credit risk forecasting task compared with the recent methods.

关键词： Training Reinforcement learning Data models Data mining Forecasting Task analysis Financial services

VideoLT: Large-scale long-tailed video recognition

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhang, Xing Wu, Zuxuan Weng, Zejia Fu, Huazhu Chen, Jingjing Jiang, Yu-Gang Davis, Larry Academy for Engineering and Technology Fudan University Shanghai Key Lab of Intel. Info. Processing School of Computer Science Fudan University Shanghai Collaborative Innovation Center on Intelligent Visual Computing Inception Institute of Artificial Intelligence University of Maryland

关键词： Image recognition

Interpreting Object-level Foundation Models via Visual Precision Search

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Ruoyu Liang, Siyuan Li, Jingzhi Liu, Shiming Li, Maosen Huang, Zhen Zhang, Hua Cao, Xiaochun Institute of Information Engineering Chinese Academy of Sciences Beijing100093 China School of Cyber Security University of Chinese Academy of Sciences Beijing100049 China School of Computing NUS Singapore RAMS Lab Huawei Technologies Co. Ltd. China IAS BU Huawei Technologies Co. Ltd. China College of Computer NUDT China Key Lab. of Edu. Inf. for Nationalities YNNU Ministry of Education Kunming China School of Cyber Science and Technology Sun Yat-sen University Shenzhen Campus Shenzhen518107 China

Advances in multimodal pre-training have propelled object-level foundation models, such as Grounding DINO and Florence-2, in tasks like visual grounding and object detection. However, interpreting these models’ decisions has grown increasingly challenging. Existing interpretable attribution methods for object-level task interpretation have notable limitations: (1) gradient-based methods lack precise localization due to visual-textual fusion in foundation models, and (2) perturbation-based methods produce noisy saliency maps, limiting fine-grained interpretability. To address these, we propose a Visual Precision Search method that generates accurate attribution maps with fewer regions. Our method bypasses internal model parameters to overcome attribution issues from multimodal fusion, dividing inputs into sparse sub-regions and using consistency and collab.ration scores to accurately identify critical decision-making regions. We also conducted a theoretical analysis of the boundary guarantees and scope of applicability of our method. Experiments on RefCOCO, MS COCO, and LVIS show our approach enhances object-level task interpretability over SOTA for Grounding DINO and Florence-2 across various evaluation metrics, with faithfulness gains of 23.7%, 31.6%, and 20.1% on MS COCO, LVIS, and RefCOCO for Grounding DINO, and 50.7% and 66.9% on MS COCO and RefCOCO for Florence-2. Additionally, our method can interpret failures in visual grounding and object detection tasks, surpassing existing methods across multiple evaluation metrics. The code is released at https://***/RuoyuChen10/VPS. Copyright © 2024, The Authors. All rights reserved.

关键词： Object detection

Hard instance learning for quantum adiabatic prime factorization

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Lin, Jian Zhang, Zhengfeng Zhang, Junping Li, Xiaopeng State Key Laboratory of Surface Physics Institute of Nanoelectronics and Quantum Computing Department of Physics Fudan University Shanghai200433 China Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai200433 China Shanghai Qi Zhi Institute AI Tower Xuhui District Shanghai200232 China

Prime factorization is a difficult problem with classical computing, whose exponential hardness is the foundation of Rivest-Shamir-Adleman (RSA) cryptography. With programmable quantum devices, adiabatic quantum computing has been proposed as a plausible approach to solve prime factorization, having promising advantage over classical computing. Here, we find there are certain hard instances that are consistently intractable for both classical simulated annealing and un-configured adiabatic quantum computing (AQC). Aiming at an automated architecture for optimal configuration of quantum adiabatic factorization, we apply a deep reinforcement learning (RL) method to configure the AQC algorithm. By setting the success probability of the worst-case problem instances as the reward to RL, we show the AQC performance on the hard instances is dramatically improved by RL configuration. The success probability also becomes more evenly distributed over different problem instances, meaning the configured AQC is more stable as compared to the un-configured case. Through a technique of transfer learning, we find prominent evidence that the framework of AQC configuration is scalab.e—the configured AQC as trained on five qubits remains working efficiently on nine qubits with a minimal amount of additional training cost. Copyright © 2021, The Authors. All rights reserved.

关键词： Factorization

Accurate Latent Factor Analysis via Particle Swarm Optimizers

学校读者我要写书评

暂无评论

Accurate Latent Factor Analysis via Particle Swarm Optimizer...

IEEE International Conference on Systems, Man and Cybernetics

作者： Jia Chen Xin Luo MengChu Zhou School of Cyber Science and Technology Beihang University Beijing China Chongqing Key Lab. of Big Data and Intelligent Computing and the Chongqing Engineering Research Center of Big Data Application for Smart Cities Chongqing Institute of Green and Intelligent Technology Chongqing School University of Chinese Academy of Sciences Chongqing China New Jersey Institute of Technology Newark NJ USA

ISBN: (纸本)9781665442084

A stochastic-gradient-descent-based Latent Factor Analysis (LFA) model is highly efficient in representative learning of a High-Dimensional and Sparse (HiDS) matrix. Its learning rate adaptation is vital in ensuring its efficiency. Such adaptation can be realized with an evolutionary computing algorithm. However, a resultant model tends to suffer from two issues: a) the pre-mature convergence of the swarm of learning rates as caused by an adopted evolution algorithm, and b) the pre-mature convergence of the LFA model as caused jointly by evolution-based learning rate adaptation and an optimization algorithm. This paper focuses on the methods to address such issues. A Hierarchical Particle-swarm-optimization-incorporated Latent factor analysis (HPL) model with a two-layered structure is proposed, where the first layer pre-trains desired latent factors with a position-transitional particle-swarm-optimization-based LFA model, and the second layer performs latent factor refining with a newly-proposed mini-batch particle swarm optimizer. With such design, an HPL model can well handle the pre-mature convergence, which is supported by the positive experimental results achieved on HiDS matrices from industrial applications.

关键词： Adaptation models Analytical models Computational modeling Refining Predictive models Sparse matrices Particle swarm optimization

CDL: Curriculum Dual Learning for Emotion-Controllab.e Response Generation

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Shen, Lei Feng, Yang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China

Emotion-controllab.e response generation is an attractive and valuable task that aims to make open-domain conversations more empathetic and engaging. Existing methods mainly enhance the emotion expression by adding regularization terms to standard cross-entropy loss and thus influence the training process. However, due to the lack of further consideration of content consistency, the common problem of response generation tasks, safe response, is intensified. Besides, query emotions that can help model the relationship between query and response are simply ignored in previous models, which would further hurt the coherence. To alleviate these problems, we propose a novel framework named Curriculum Dual Learning (CDL) which extends the emotion-controllab.e response generation to a dual task to generate emotional responses and emotional queries alternatively. CDL utilizes two rewards focusing on emotion and content to improve the duality. Additionally, it applies curriculum learning to gradually generate high-quality responses based on the difficulties of expressing various emotions. Experimental results show that CDL significantly outperforms the baselines in terms of coherence, diversity, and relation to emotion factors. Copyright © 2020, The Authors. All rights reserved.

关键词： Curricula

DMNER: Biomedical Named Entity Recognition by Detection and Matching

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Bian, Junyi Jiang, Rongze Zhai, Weiqi Huang, Tianyang Zhou, Hong Zhu, Shanfeng School of Computer Science Fudan University Shanghai200433 China Atypon Systems LLC United Kingdom Institute of Science and Technology for Brain-Inspired Intelligence Fudan University China Key Laboratory of Computational Neuroscience and Brain-Inspired Intelligence Fudan University Ministry of Education Shanghai200433 China MOE Frontiers Center for Brain Science Fudan University Shanghai200433 China Zhangjiang Fudan International Innovation Center Shanghai200433 China Shanghai Key Lab of Intelligent Information Processing Fudan University Shanghai200433 China

Biomedical named entity recognition (BNER) serves as the foundation for numerous biomedical text mining tasks. Unlike general NER, BNER require a comprehensive grasp of the domain, and incorporating external knowledge beyond training data poses a significant challenge. In this study, we propose a novel BNER framework called DMNER. By leveraging existing entity representation models SAP-BERT, we tackle BNER as a two-step process: entity boundary detection and biomedical entity matching. DMNER exhibits applicability across multiple NER scenarios: 1) In supervised NER, we observe that DMNER effectively rectifies the output of baseline NER models, thereby further enhancing performance. 2) In distantly supervised NER, combining MRC and AutoNER as span boundary detectors enables DMNER to achieve satisfactory results. 3) For training NER by merging multiple datasets, we adopt a framework similar to DS-NER but additionally leverage ChatGPT to obtain high-quality phrases in the training. Through extensive experiments conducted on 10 benchmark datasets, we demonstrate the versatility and effectiveness of DMNER. Copyright © 2023, The Authors. All rights reserved.

关键词： Character recognition

ISIA food-500: A dataset for large-scale food recognition via stacked global-local attention network

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Min, Weiqing Liu, Linhu Wang, Zhiling Luo, Zhengdong Wei, Xiaoming Wei, Xiaolin Jiang, Shuqiang Key Lab of Intelligent Information Processing Institute of Computing Technology CAS Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China Meituan-Dianping Group China

Food recognition has received more and more attention in the multimedia community for its various real-world applications, such as diet management and self-service restaurants. A large-scale ontology of food images is urgently needed for developing advanced large-scale food recognition algorithms, as well as for providing the benchmark dataset for such algorithms. To encourage further progress in food recognition, we introduce the dataset ISIA Food-500 with 500 categories from the list in the Wikipedia and 399,726 images, a more comprehensive food dataset that surpasses existing popular benchmark datasets by category coverage and data volume. Furthermore, we propose a stacked global-local attention network, which consists of two sub-networks for food recognition. One sub-network first utilizes hybrid spatial-channel attention to extract more discriminative features, and then aggregates these multi-scale discriminative features from multiple layers into global-level representation (e.g., texture and shape information about food). The other one generates attentional regions (e.g., ingredient relevant regions) from different regions via cascaded spatial transformers, and further aggregates these multi-scale regional features from different layers into local-level representation. These two types of features are finally fused as comprehensive representation for food recognition. Extensive experiments on ISIA Food-500 and other two popular benchmark datasets demonstrate the effectiveness of our proposed method, and thus can be considered as one strong baseline. The dataset, code and models can be found at http://123.57.42.89/Foodcomputing-Dataset/***. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep learning