检索结果-内蒙古大学图书馆

20th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2023

作者： Li, Weitao Li, Shijun Shang, Lin State Key Laboratory for Novel Software Technology Nanjing University Nanjing210023 China Department of Computer Science and Technology Nanjing University Nanjing210023 China

ISBN: (纸本)9789819970247

The susceptibility of current Deep Neural Networks (DNNs) to adversarial examples has been a significant concern in deep learning methods. In particular, sparse adversarial examples represent a specific category of adversarial examples that can deceive the target model by perturbing only a few pixels in images. While existing sparse adversarial attack methods have shown achievements, the current results of sparsity and efficiency are inadequate and require significant improvements. This paper introduces an adv-triplet loss and proposes a search attack method to attack the Face Expression Recognition (FER) system with minimal pixel perturbations. Specifically, we propose an adv-triplet loss function and utilize its gradient information to generate pixels for adversarial examples. Extensive experiments conducted on the CK+ and Oulu-CASIA datasets demonstrate the superiority of our proposed method over several state-of-the-art sparse attack methods. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2024.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

DOS: DIVERSE OUTLIER SAMPLING FOR OUT-OF-DISTRIBUTION DETECTION 12

DOS: DIVERSE OUTLIER SAMPLING FOR OUT-OF-DISTRIBUTION DETECT...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Jiang, Wenyu Cheng, Hao Chen, Mingcai Wang, Chongjun Wei, Hongxin Department of Statistics and Data Science Southern University of Science and Technology China State Key Laboratory for Novel Software Technology Nanjing University China

Modern neural networks are known to give overconfident predictions for out-of-distribution inputs when deployed in the open world. It is common practice to leverage a surrogate outlier dataset to regularize the model during training, and recent studies emphasize the role of uncertainty in designing the sampling strategy for outlier datasets. However, the OOD samples selected solely based on predictive uncertainty can be biased towards certain types, which may fail to capture the full outlier distribution. In this work, we empirically show that diversity is critical in sampling outliers for OOD detection performance. Motivated by the observation, we propose a straightforward and novel sampling strategy named DOS (Diverse Outlier Sampling) to select diverse and informative outliers. Specifically, we cluster the normalized features at each iteration, and the most informative outlier from each cluster is selected for model training with absent category loss. With DOS, the sampled outliers efficiently shape a globally compact decision boundary between ID and OOD data. Extensive experiments demonstrate the superiority of DOS, reducing the average FPR95 by up to 25.79% on CIFAR-100 with TI-300K. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Statistics

来源：评论

学校读者我要写书评

暂无评论

Knowledge Graph Driven Inference Testing for Question Answering software 24

Knowledge Graph Driven Inference Testing for Question Answer...

引用

44th ACM/IEEE International Conference on software Engineering, ICSE 2024

作者： Wang, Jun Li, Yanhui Chen, Zhifei Chen, Lin Zhang, Xiaofang Zhou, Yuming Nanjing University State Key Laboratory for Novel Software Technology China School of Computer Science and Engineering Nanjing University of Science and Technology China School of Computer Science and Technology Soochow University China

ISBN: (纸本)9798400702174

In the wake of developments in the field of Natural Language Processing, Question Answering (QA) software has penetrated our daily lives. Due to the data-driven programming paradigm, QA software inevitably contains bugs, i.e., misbehaving in real-world applications. Current testing techniques for testing QA software include two folds, reference-based testing and metamorphic testing. This paper adopts a different angle to achieve testing for QA software: we notice that answers to questions would have inference relations, i.e., the answers to some questions could be logically inferred from the answers to other questions. If these answers on QA software do not satisfy the inference relations, an inference bug is detected. To generate the questions with the inference relations automatically, we propose a novel testing method Knowledge Graph driven Inference Testing (KGIT), which employs facts in the Knowledge Graph (KG) as the seeds to logically construct test cases containing questions and contexts with inference relations. To evaluate the effectiveness of KGIT, we conduct an extensive empirical study with more than 2.8 million test cases generated from the large-scale KG YAGO4 and three QA models based on the state-of-the-art QA model structure. The experimental results show that our method (a) could detect a considerable number of inference bugs in all three studied QA models and (b) is helpful in retraining QA models to improve their inference ability. © 2024 ACM.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

PredRacer: Predictively Detecting Data Races in Android Applications 31

PredRacer: Predictively Detecting Data Races in Android Appl...

引用

31st IEEE International Conference on software Analysis, Evolution and Reengineering, SANER 2024

作者： Guo, Xin Qi, Xiaofang Li, Yanhui Wu, Chao School of Computer Science and Engineering Southeast University China Nanjing University State Key Laboratory for Novel Software Technology China

ISBN: (纸本)9798350330663

Android platform offers a hybrid concurrency model encompassing multi-threading and asynchronous messaging for concurrent programming. The model is powerful but complex, making it difficult for developers to analyze concurrent behaviors. Data race, a prevalent concurrency defect in real-world Android applications, often results in abnormal executions of mobile applications, even crashes. Despite quite a few studies on detecting data races, it still suffers from high false positives with static analysis techniques and high false negatives with dynamic analysis techniques. To address this issue, this paper presents a predictive approach, PredRacer, for detecting data races in Android applications. It first captures an execution trace of an Android application, and then reorders the events within the trace based on partial orders. Finally, it checks the feasibility of the generated event sequence, which contains a potential data race. PredRacer increases the search scope and reduces false negatives while mitigating false positives by incorporating the happen-before relations specific to the Android concurrency model. The effectiveness of PredRacer is evaluated using the BenchERoid data set. Experimental results demonstrate that PredRacer achieves high precision, recall, and F1 score, outperforming the state-of-the-art techniques. A collection of 20 open-source Android applications is further utilized to assess the effectiveness of PredRacer, and an evaluation of 300 wild apps is conducted to assess its efficiency and scalability. © 2024 IEEE.

关键词： Static analysis

来源：评论

学校读者我要写书评

暂无评论

USee: Ultrasound-based Device-free Eye Movement Sensing 21

USee: Ultrasound-based Device-free Eye Movement Sensing

引用

21st Annual IEEE International Conference on Sensing, Communication, and Networking, SECON 2024

作者： Cheng, Wen Pang, Mingzhi Wan, Haoran Dong, Shichen Liu, Dongxu Wang, Wei State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Department of Computer Science Princeton University Princeton United States

ISBN: (纸本)9798331519186

Eye movements play a significant role in human-computer interaction and are widely recognized as an essential health indicator, making their detection both appealing and technically challenging. In this paper, we present a system named USEE that achieves high-precision capture of weak and aperiodic eye movements by utilizing fine-grained and ubiquitous ultrasound signals, capturing both blinking and more subtle saccades. We first identify signal changes associated with eye movements by capturing the unique impact of blinking. Further, we establish a pioneering relationship between the residuals from signal decomposition and subtle eye movements. Utilizing innovative signal processing architectures, we mitigate interference and effectively extract eye movement features. Subsequently, we employ one-dimensional convolutional operations in place of signal cross-correlation, designing filters for motion category identification and a lightweight convolutional neural network for saccade direction classification. This enables our system to serve as a foundational sensing layer for eye movement tracking, applicable across diverse applications. We implement USEE on both a research-purpose platform and a commodity Raspberry Pi. Extensive experimental results demonstrate the effectiveness of our system, achieving 91% accuracy in saccade recognition and 94% in blink detection. The system proves robust, even in challenging scenarios with strong interference, such as the presence of moving pedestrians. © 2024 IEEE.

关键词： Motion sensors

来源：评论

学校读者我要写书评

暂无评论

Property Constrained Video Summarization via Regret Minimization

引用

SN computer science 2024年第2期5卷 254页

作者： Xu, Yuyao Zheng, Jiping Tao, Yanxin Zhu, Kaiqin College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing 211106 China State Key Laboratory for Novel Software Technology Nanjing University Nanjing 210023 China Collaborative Innovation Center of Novel Software Technology and Industrialization Nanjing 210016 China

Video summarization has become one of the most effective solutions for quickly understanding a large amount of video data. Video properties such as importance, diversity, representativeness, and storyness have been widely adopted for summarization based on kinds of features of video frames. To fully exploit these properties, in this paper, we propose a property constrained video summarization framework to output fixed-size summaries based on the concept of regret minimization which is popular in the database community for solving multi-criteria decision-making problems. Our framework first uses representativeness property and effective techniques to generate a candidate frame set. Then, the candidate set is transformed into a multi-dimensional point set where dimensions are composed of entities of the importance property and the diversity property. Moreover, our framework runs the regret minimization query on the point set to output a summary. To further improve the quality of the output summary, the property storyness is exploited to add frames satisfying the size constraint. Experimental results on the benchmark datasets demonstrate the effectiveness and superiority of our framework. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd 2024.

关键词： Multiple properties Regret minimization Video summarization

来源：评论

学校读者我要写书评

暂无评论

Grey-box Adversarial Attack on Communication in Multi-agent Reinforcement Learning 22

Grey-box Adversarial Attack on Communication in Multi-agent ...

引用

22nd International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2023

作者： Ma, Xiao Li, Wu-Jun National Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University Nanjing China

Although research on communication in multi-agent reinforcement learning (MARL) has achieved some progress, the vulnerability of the communication mechanism in MARL caused by adversarial communication messages generated by malicious agents has not been well investigated. Existing works about adversarial communication messages in MARL focus on the black-box scenario where the attacker cannot access any model information about the multi-agent system (MAS). But a more practical setting is the grey-box scenario where the attacker can access the model information about its controlled agent. To the best of our knowledge, there has not been any work investigating grey-box attacks on communication in MARL. In this paper, we propose the first grey-box attack method on communication in MARL, which is called victim-simulation based adversarial attack (VSA). At each timestep, the attacker simulates a victim attacked by other regular agents' communication messages and generates adversarial perturbations on its received communication messages. The aggregation of these perturbations is sent by the attacker to the regular agents through communication messages, which will induce non-optimal actions of the regular agents. Experimental results show that VSA can effectively degrade the performance of the MAS on Predator-Prey. The findings in this paper will make researchers aware of the grey-box attack in MARL. © 2023 International Foundation for Autonomous Agents and Multiagent Systems (***). All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

RandoMix: a mixed sample data augmentation method with multiple mixed modes

引用

Multimedia Tools and Applications 2025年第8期84卷 4343-4359页

作者： Liu, Xiaoliang Shen, Furao Zhao, Jian Nie, Changhai National Key Laboratory for Novel Software Technology Nanjing University Nanjing China Department of Computer Science and Technology Nanjing University Nanjing China School of Artificial Intelligence Nanjing University Nanjing China School of Electronic Science and Engineering Nanjing University Nanjing China

Data augmentation plays a crucial role in enhancing the robustness and performance of machine learning models across various domains. In this study, we introduce a novel mixed-sample data augmentation method called RandoMix. RandoMix is specifically designed to simultaneously address robustness and diversity challenges. It leverages a combination of linear and mask mixed modes, introducing flexibility in candidate selection and weight adjustments. We evaluate the effectiveness of RandoMix on diverse datasets, including CIFAR-10/100, Tiny-ImageNet, ImageNet, and Google Speech Commands. Our results demonstrate its superior erformance compared to existing techniques such as Mixup, CutMix, Fmix, and ResizeMix. Notably, RandoMix excels in enhancing model robustness against adversarial noise, natural noise, and sample occlusion. The comprehensive experimental results and insights into parameter tuning underscore the potential of RandoMix as a versatile and effective data augmentation method. Moreover, it seamlessly integrates into the training pipeline. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Data augmentation Diversity Mixed-sample RandoMix Robustness

来源：评论

学校读者我要写书评

暂无评论

Autonomous Aspect-Image Instruction A2II: Q-Former Guided Multimodal Sentiment Classification 30

Autonomous Aspect-Image Instruction A2II: Q-Former Guided Mu...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Feng, Junjia Lin, Mingqian Shang, Lin Gao, Xiaoying State Key Laboratory for Novel Software Technology Nanjing University China Department of Computer Science and Technology Nanjing University China School of Engineering and Computer Science Victoria University of Wellington New Zealand

ISBN: (纸本)9782493814104

Multimodal aspect-oriented sentiment classification (MABSC) task has garnered significant attention, which aims to identify the sentiment polarities of aspects by combining both language and vision information. However, the limited multimodal data in this task has become a big gap for the vision-language multimodal fusion. While large-scale vision-language pretrained models have been adapted to multiple tasks, their use for MABSC task is still in a nascent stage. In this work, we present an attempt to use the instruction tuning paradigm to MABSC task and leverage the ability of large vision-language models to alleviate the limitation in the fusion of textual and image modalities. To tackle the problem of potential irrelevance between aspects and images, we propose a plug-and-play selector to autonomously choose the most appropriate instruction from the instruction pool, thereby reducing the impact of irrelevant image noise on the final sentiment classification results. We conduct extensive experiments in various scenarios and our model achieves state-of-the-art performance on benchmark datasets, as well as in few-shot settings. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Self-Training with Label-Feature-Consistency for Domain Adaptation 28th

Self-Training with Label-Feature-Consistency for Domain Ad...

引用

28th International Conference on Database Systems for Advanced Applications, DASFAA 2023

作者： Xin, Yi Luo, Siqi Jin, Pengsheng Du, Yuntao Wang, Chongjun National Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University Nanjing China

ISBN: (纸本)9783031306778

Mainstream approaches for unsupervised domain adaptation (UDA) learn domain-invariant representations to address the domain shift. Recently, self-training has been used in UDA, which exploits pseudo-labels for unlabeled target domains. However, the pseudo-labels can be unreliable due to distribution shifts between domains, severely impairing the model performance. To address this problem, we propose a novel self-training framework-Self-Training with Label-Feature-Consistency (ST-LFC), which selects reliable target pseudo-labels via label-level and feature-level voting consistency principle. The former means target pseudo-labels generated by a source-trained classifier and the latter means the nearest source-class to the target in feature space. In addition, ST-LFC reduces the negative effects of unreliable predictions through entropy minimization. Empirical results indicate that ST-LFC significantly improves over the state-of-the-arts on a variety of benchmark datasets. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Entropy

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：