检索结果-内蒙古大学图书馆

Improving Multi-Label Contrastive Learning by Leveraging Label Distribution

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Chen, Ning Lyu, Shen-Huan Wu, Tian-Shuang Wang, Yanyan Tang, Bin Key Laboratory of Water Big Data Technology of Ministry of Water Resources College of Computer Science and Software Engineering Hohai University Nanjing China State Key Laboratory for Novel Software Technology Nanjing University Nanjing China

In multi-label learning, leveraging contrastive learning to learn better representations faces a key challenge: selecting positive and negative samples and effectively utilizing label information. Previous studies selected positive and negative samples based on the overlap between labels and used them for label-wise loss balancing. However, these methods suffer from a complex selection process and fail to account for the varying importance of different labels. To address these problems, we propose a novel method that improves multi-label contrastive learning through label distribution. Specifically, when selecting positive and negative samples, we only need to consider whether there is an intersection between labels. To model the relationships between labels, we introduce two methods to recover label distributions from logical labels, based on Radial Basis Function (RBF) and contrastive loss, respectively. We evaluate our method on nine widely used multi-label datasets, including image and vector datasets. The results demonstrate that our method outperforms state-of-the-art methods in six evaluation metrics. Copyright © 2025, The Authors. All rights reserved.

关键词： Contrastive Learning

Compressing Model with Few Class-Imbalance Samples: An Out-of-Distribution Expedition

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wu, Tian-Shuang Lyu, Shen-Huan Chen, Ning Qu, Zhihao Ye, Baoliu Key Laboratory of Water Big Data Technology Ministry of Water Resources College of Computer Science and Software Engineering Hohai University Nanjing China State Key Laboratory for Novel Software Technology Nanjing University Nanjing China

In recent years, as a compromise between privacy and performance, few-sample model compression has been widely adopted to deal with limited data resulting from privacy and security concerns. However, when the number of available samples is extremely limited, class imbalance becomes a common and tricky problem. Achieving an equal number of samples across all classes is often costly and impractical in real-world applications, and previous studies on few-sample model compression have mostly ignored this significant issue. Our experiments comprehensively demonstrate that class imbalance negatively affects the overall performance of few-sample model compression methods. To address this problem, we propose a novel and adaptive framework named OOD-Enhanced Few-Sample Model Compression (OE-FSMC). This framework integrates easily accessible out-of-distribution (OOD) data into both the compression and fine-tuning processes, effectively rebalancing the training distribution. We also incorporate a joint distillation loss and a regularization term to reduce the risk of the model overfitting to the OOD data. Extensive experiments on multiple benchmark datasets show that our framework can be seamlessly incorporated into existing few-sample model compression methods, effectively mitigating the accuracy degradation caused by class imbalance. Copyright © 2025, The Authors. All rights reserved.

关键词： Differential privacy

A Logical Pattern Memory Pre-trained Model for Entailment Tree Generation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yuan, Li Cai, Yi Ren, Haopeng Wang, Jiexin School of Software Engineering South China University of Technology China Key Laboratory of Big Data and Intelligent Robot South China University of Technology Ministry of Education China

Generating coherent and credible explanations remains a significant challenge in the field of AI. In recent years, researchers have delved into the utilization of entailment trees to depict explanations, which exhibits a reasoning process of how a hypothesis is deduced from the supporting facts. However, existing models often overlook the importance of generating intermediate conclusions with logical consistency from the given facts, leading to inaccurate conclusions and undermining the overall credibility of entailment trees. To address this limitation, we propose the logical pattern memory pre-trained model (LMPM). LMPM incorporates an external memory structure to learn and store the latent representations of logical patterns, which aids in generating logically consistent conclusions. Furthermore, to mitigate the influence of logically irrelevant domain knowledge in the Wikipedia-based data, we introduce an entity abstraction approach to construct the dataset for pre-training LMPM. The experimental results highlight the effectiveness of our approach in improving the quality of entailment tree generation. By leveraging logical entailment patterns, our model produces more coherent and reasonable conclusions that closely align with the underlying premises. Code and data are released at https://***/YuanLi95/T5-LMPM. © 2024, CC BY-NC-ND.

关键词： Domain Knowledge

Neighbor Relations Matter in Video Scene Detection

学校读者我要写书评

暂无评论

Neighbor Relations Matter in Video Scene Detection

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Jiawei Tan Hongxing Wang Jiaxin Li Zhilong Ou Zhangbin Qian Key Laboratory of Dependable Service Computing in Cyber Physical Society (Chongqing University) Ministry of Education China School of Big Data and Software Engineering Chongqing University China

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Video scene detection aims to temporally link shots for obtaining semantically compact scenes. It is essential for this task to capture scene-distinguishable affinity among shots by similarity assessment. However, most methods relies on ordinary shot-to-shot similarities, which may inveigle similar shots into being linked even though they are from different scenes, and meanwhile hinder dissimilar shots from being blended into a complete scene. In this paper, we propose NeighborNet to inject shot contexts into shot-to-shot similarities through carefully exploring the relations between semantic/temporal neighbors of shots over a local time period. In this way, shot-to-shot similarities are remeasured as semantic/temporal neighbor-aware similarities so that NeighborNet can learn context embedding into shot features using graph convolutional network. As a result, not only do the learned shot features suppress the affinity among similar shots from different scenes, but they also promote the affinity among dissimilar shots in the same scene. Experimental results on public benchmark datasets show that our proposed NeighborNet yields substantial improvements in video scene detection, especially outperforms released state-of-the-arts by at least 6% in Average Precision (AP). The code is available at https://***/ExMorgan-Alter/NeighborNet.

关键词： Learning systems Computer vision Codes Graph convolutional networks Image edge detection Semantics Benchmark testing

An Alternative Algorithm for the Symmetry Classification of Ordinary Differential Equations

学校读者我要写书评

暂无评论

Sound & Vibration 2022年第1期56卷 65-76页

作者： Yi Tian Jing Pang College of Data Science and Application Inner Mongolia University of TechnologyHohhot010080China Inner Mongolia Autonomous Region Engineering and Technology Research Center of Big Data Based Software Service Hohhot010080China College of Science Inner Mongolia University of TechnologyHohhot010051China Inner Mongolia Key Laboratory of Statistical Analysis Theory for Life Data and Neural Network Modeling Inner Mongolia University of TechnologyHohhot010051China

This is the first paper on symmetry classification for ordinary differential equations(ODEs)based on Wu’s *** carry out symmetry classification of two ODEs,named the generalizations of the Kummer-Schwarz equations which involving arbitrary ***,Lie algorithm is used to give the determining equations of symmetry for the given equations,which involving arbitrary ***,differential form Wu’s method is used to decompose determining equations into a union of a series of zero sets of differential characteristic sets,which are easy to be solved *** branch of the decomposition yields a class of symmetries and associated *** algorithm makes the classification become direct and *** Dimitrov Bozhkov,and Pammela Ramos da Conceição have used the Lie algorithm to give the symmetry classifications of the equations talked in this paper in *** this paper,we can find that the differential form Wu’s method for symmetry classification of ODEs with arbitrary function(parameter)is effective,and is an alternative method.

关键词： Kummer-Schwarz equation ordinary differential equations(ODEs) differential form Wu’s method

Computing Salient Feature Points of 3D Model Based on Geodesic Distance and Decision Graph Clustering

学校读者我要写书评

暂无评论

Recent Advances in Computer Science and Communications 2021年第8期14卷 2489-2494页

作者： Sun, Dechao Chen, Nenglun Wang, Renfang Liu, Bangquan Liang, Feng College of Big Data and Software Engineering Zhejiang Wanli University Ningbo China Department of Computer Science Hong Kong University Hong Kong College of Digital Technology and Engineering Ningbo University of Finance & Economics Ningbo China

Introduction: Computing Salient Feature Points (SFP) of 3D models has important application value in the field of computer graphics. In order to extract the SFP more effectively, a novel SFP computing algorithm based on geodesic distance and decision graph clustering is proposed. Methods: Firstly, the geodesic distance of model vertices is calculated based on the heat conduction equation, and then the average geodesic distance and importance weight of vertices are calculated. Finally, the decision graph clustering method is used to calculate the decision graph of model vertices. Results and Discussion: 3D models in SHREC 2011 dataset are selected to test the proposed algorithm. Compared with the existing algorithms, this method calculates the SFP of the 3D model from a global perspective. Results show that it is not affected by model posture and noise. Conclusion: Our method maps the SFP of the 3D model to the 2D decision-making diagram, which simplifies the calculation process of SFP, improves the calculation accuracy and possesses strong robustness. © 2021 Bentham Science Publishers.

关键词： Clustering algorithms

Spatial-Semantic Collaborative Cropping for User Generated Content

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Su, Yukun Cao, Yiwen Deng, Jingliang Rao, Fengyun Wu, Qingyao School of Software and Engineering South China University of Technology China WeChat Tencent Inc China Key Laboratory of Big Data and Intelligent Robot Ministry of Education China

A large amount of User Generated Content (UGC) is uploaded to the Internet daily and displayed to people world-widely through the client side (e.g., mobile and PC). This requires the cropping algorithms to produce the aesthetic thumbnail within a specific aspect ratio on different devices. However, existing image cropping works mainly focus on landmark or landscape images, which fail to model the relations among the multi-objects with the complex background in UGC. Besides, previous methods merely consider the aesthetics of the cropped images while ignoring the content integrity, which is crucial for UGC cropping. In this paper, we propose a Spatial-Semantic Collaborative cropping network (S2CNet) for arbitrary user generated content accompanied by a new cropping benchmark. Specifically, we first mine the visual genes of the potential objects. Then, the suggested adaptive attention graph recasts this task as a procedure of information association over visual nodes. The underlying spatial and semantic relations are ultimately centralized to the crop candidate through differentiable message passing, which helps our network efficiently to preserve both the aesthetics and the content integrity. Extensive experiments on the proposed UGCrop5K and other public datasets demonstrate the superiority of our approach over state-of-the-art counterparts. Our project is available at https://***/suyukun666/S2CNet. Copyright © 2024, The Authors. All rights reserved.

关键词： Message passing

Task Scheduling in Server Farms with Reinforcement Learning

学校读者我要写书评

暂无评论

Task Scheduling in Server Farms with Reinforcement Learning

Computer Information Science and Application Technology (CISAT), International Conference on

作者： Chao Deng Jianxiong Wan Chuyi Liu Leixiao Li Inner Mongolia University of Technology Huhhot China Inner Mongolia Autonomous Region Engineering & Technology Research Center of Big Data Based Software Service Huhhot China

ISBN: (数字)9798350375107

ISBN: (纸本)9798350375114

Multi-task concurrent scheduling is crucial to improving system performance. Commonly used traditional scheduling algorithms (random, greedy, etc.) are usually difficult to cope with dynamic tasks and resource requirements, resulting in unsatisfactory results. This paper proposes a deep reinforcement learning (DRL) method to solve the scheduling problem by modeling it as a Markov decision process (MDP). The Deep Q Network (DQN) algorithm uses experience replay to enhance learning stability and performance through training strategies and value networks. The training reward results show that it is 40.1 7% higher than the random scheduling strategy, proving its effectiveness.

关键词： Training Greedy algorithms Scheduling algorithms Heuristic algorithms System performance Multitasking Stability analysis Resource management Servers Optimization

CLIP-Driven Multi-Scale Instance Learning for Weakly Supervised Video Anomaly Detection

学校读者我要写书评

暂无评论

CLIP-Driven Multi-Scale Instance Learning for Weakly Supervi...

IEEE International Conference on Multimedia and Expo (ICME)

作者： Zhangbin Qian Jiawei Tan Zhilong Ou Hongxing Wang School of Big Data and Software Engineering Chongqing University China Key Laboratory of Dependable Service Computing in Cyber Physical Society Chongqing University Ministry of Education China

ISBN: (数字)9798350390155

ISBN: (纸本)9798350390162

Existing weakly supervised video anomaly detection methods mainly employ Multiple Instance Learning (MIL) to identify abnormal snippets in untrimmed videos. However, the semantics and presentations of anomalies frequently exhibit ambiguity that MIL is difficult to tackle. Moreover, MIL suffers from false alarms due to its independent optimization of each instance, neglecting temporal correlation between adjacent snippets. Consequently, we badly need to better connect abnormal presentations and their semantics, as well as to enable multi-temporal-scale anomaly discovery. This paper proposes a CLIP-Driven Multi-Scale Instance Learning (CMSIL) framework with two branches including Vision-Language (VL) and Multi-Scale Instance Learning (MSIL). The VL branch leverages the powerful visual concept priors from Contrastive Language-Image Pre-training (CLIP) to generate pseudo anomalies, thereby providing suspected anomaly cues for model training guidance. The MSIL branch utilizes a feature pyramid to fully mine fine-grained temporal dependencies by employing MIL within each pyramid level to learn anomalous patterns across different temporal scales. By collaborating with the two branches, the proposed CMSIL shows better proficiency in handling anomalies with varying durations. Extensive experiments on the XD-Violence and UCF-Crime datasets demonstrate the superior performance of our method. The code is available at https://***/casperZB/CMSIL.

关键词： Training Visualization Correlation Codes Semantics Focusing Streaming media Anomaly detection Optimization Context modeling