检索结果-内蒙古大学图书馆

SSRN 2023年

作者： Lu, Yantao Liu, Ning Li, Yilan Chen, Jinchao Velipasalar, Senem Northwestern Polytechnical University School of Computer Science Shaanxi Xi’an China Midea Group AI Innovation Center Beijing China Xi’an University of Technology School of Computer Science and Engineering Shaanxi Xi’an China Syracuse University Department of Electrical Engineering and Computer Science NY13244 United States

Despite the rapid advances in adversarial machine learning, state-of-the-art attack methods encounter practical limitations in the field of onboard perception that require real-time and multi-task processing. Conventional attacks target on a specific perception task such as object detection or segmentation, which is difficult to simultaneously penetrate the entire multi-task perception module. Moreover, the real-time processing requirement further restrict the existing attacks to be deployed on practical autonomous driving platforms. To address these limitations, we propose Perception Streaming Attack (PSA), which is a sequence-based cross-task adversarial attack framework. We firstly propose Priori Perturbation Generator (PPG) to calculate a priori perturbation by leveraging the perturbation of previous frame as well as the motion information between the previous and current frames. Then, we propose Posterior Perturbation Updater (PPU) to refine the priori perturbation and obtain the final adversarial example for current frame. Comprehensive experimental evaluations on BDD100k and NuImages datasets demonstrate that the proposed PSA, compared with the state-of-the-art attacks, can effectively and efficiently attack across different tasks used in onboard perception. We also deploy our Perception Streaming Attack framework on a single-board computer (NVIDIA Jetson AGX Xavier) to validate the on-board performance. The experimental results show that the proposed PSA can successfully run at 12 Hz and effectively erase at least 76% objects that should be sensed. © 2023, The Authors. All rights reserved.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Be Intentional About Fairness!: Fairness, Size, and Multiplicity in the Rashomon Set

arXiv

引用

arXiv 2025年

作者： Dai, Gordon Ravishankar, Pavan Yuan, Rachel Neill, Daniel B. Black, Emily Machine Learning for Good Laboratory Department of Computer Science NYU Courant Institute of Mathematical Sciences New York University United States Robert F. Wagner Graduate School of Public Service New York University United States Center for Urban Science and Progress Tandon School of Engineering New York University United States Center for Data Science New York University United States Department of Computer Science and Engineering Tandon School of Engineering New York University United States

When selecting a model from a set of equally performant models, how much unfairness can you really reduce? Is it important to be intentional about fairness when choosing among this set, or is arbitrarily choosing among the set of "good" models good enough? Recent work has highlighted that the phenomenon of model multiplicity—where multiple models with nearly identical predictive accuracy exist for the same task—has both positive and negative implications for fairness, from strengthening the enforcement of civil rights law in ai systems to showcasing arbitrariness in ai decision-making. Despite the enormous implications of model multiplicity, there is little work that explores the properties of sets of equally accurate models, or Rashomon sets, in general. In this paper, we present five main theoretical and methodological contributions which help us to understand the relatively unexplored properties of the Rashomon set, in particular with regards to fairness. Our contributions include methods for efficiently sampling models from this set and techniques for identifying the fairest models according to key fairness metrics such as statistical parity. We also derive the probability that an individual’s prediction will be flipped within the Rashomon set, as well as expressions for the set’s size and the distribution of error tolerance used across models. These results lead to policy-relevant takeaways, such as the importance of intentionally looking for fair models within the Rashomon set, and understanding which individuals or groups may be more susceptible to arbitrary decisions. © 2025, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

CanWe Utilize Pre-trained Language Models within Causal Discovery Algorithms?

arXiv

引用

arXiv 2023年

作者： Lee, Chanhui Kim, Juhyeon Jeong, Yongjun Lyu, Juhyun Kim, Junghee Lee, Sangmin Han, Sangjun Choe, Hyeokjun Park, Soyeon Lim, Woohyung Lim, Sungbin Lee, Sanghack Department of Artificial Intelligence Korea University Korea Republic of Graduate School of Data Science Seoul National University Korea Republic of Department of Computer Science and Engineering UNIST Korea Republic of Data Intelligence Laboratory LG AI Research Korea Republic of Department of Statistics Korea University Korea Republic of LG AI Research Korea Republic of SNU-LG AI Research Center Korea Republic of

Scaling laws have allowed Pre-trained Language Models (PLMs) into the field of causal reasoning. Causal reasoning of PLM relies solely on text-based descriptions, in contrast to causal discovery which aims to determine the causal relationships between variables utilizing data. Recently, there has been current research regarding a method that mimics causal discovery by aggregating the outcomes of repetitive causal reasoning, achieved through specifically designed prompts [16]. It highlights the usefulness of PLMs in discovering cause and effect, which is often limited by a lack of data, especially when dealing with multiple variables. Conversely, the characteristics of PLMs which are that PLMs do not analyze data and they are highly dependent on prompt design leads to a crucial limitation for directly using PLMs in causal discovery. Accordingly, PLM-based causal reasoning deeply depends on the prompt design and carries out the risk of overconfidence and false predictions in determining causal relationships. In this paper, we empirically demonstrate the aforementioned limitations of PLM-based causal reasoning through experiments on physics-inspired synthetic data. Then, we propose a new framework that integrates prior knowledge obtained from PLM with a causal discovery algorithm. This is accomplished by initializing an adjacency matrix for causal discovery and incorporating regularization using prior knowledge. Our proposed framework not only demonstrates improved performance through the integration of PLM and causal discovery but also suggests how to leverage PLM-extracted prior knowledge with existing causal discovery algorithms. Copyright © 2023, The Authors. All rights reserved.

关键词： Knowledge management

来源：评论

学校读者我要写书评

暂无评论

1000 FPS HDR Video with a Spike-RGB Hybrid Camera

1000 FPS HDR Video with a Spike-RGB Hybrid Camera

引用

Conference on computer Vision and Pattern Recognition (CVPR)

作者： Yakun Chang Chu Zhou Yuchen Hong Liwen Hu Chao Xu Tiejun Huang Boxin Shi National Key Laboratory for Multimedia Information Processing School of Computer Science Peking University National Engineering Research Center of Visual Technology School of Computer Science Peking University National Key Laboratory of General AI School of Intelligence Science and Technology Peking University

Capturing high frame rate and high dynamic range (HFR&HDR) color videos in high-speed scenes with conventional frame-based cameras is very challenging. The increasing frame rate is usually guaranteed by using shorter exposure time so that the captured video is severely interfered by noise. Alternating exposures can alleviate the noise issue but sacrifice frame rate due to involving long-exposure frames. The neuromorphic spiking camera records high-speed scenes of high dynamic range without colors using a completely different sensing mechanism and visual representation. We introduce a hybrid camera system composed of a spiking and an alternating-exposure RGB camera to capture HFR&HDR scenes with high fidelity. Our insight is to bring each camera's superiority into full play. The spike frames, with accurate fast motion information encoded, are firstly reconstructed for motion representation, from which the spike-based optical flows guide the recovery of missing temporal information for long-exposure RGB images while retaining their reliable color appearances. With the strong temporal constraint estimated from spike trains, both missing and distorted colors cross RGB frames are recovered to generate time-consistent and HFR color frames. We collect a new Spike-RGB dataset that contains 300 sequences of synthetic data and 20 groups of real-world data to demonstrate 1000 FPS HDR videos outperforming HDR video reconstruction methods and commercial high-speed cameras.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Effective Training Strategy for NN Models of Working Memory Classification with Limited Samples

Effective Training Strategy for NN Models of Working Memory ...

引用

IEEE International Symposium on Biomedical Imaging

作者： Pranav Suresh Bhaskar Ray Bishal Thapaliya Britny Farahdel Behnam Kazemivash Jiayu Chen Kuaikuai Duan Vince D. Calhoun Jingyu Liu Tri-Institutional Center for Translational Research in Neuroimaging and Data Science Department of Computer Science Georgia State University Atlanta USA School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta USA

In biomedical imaging, we often encounter tasks requiring machine learning models to analyze large amounts of variables. However, in practice, limited and imbalanced samples hinder the prediction power of the machine learning methods, particularly for data-hungry techniques like Neural Networks (NN). This study proposes a strategy for training NN models with limited samples. When the data condition result in a model with suboptimal solutions even under a shallow network configuration and best learning practices, we propose a simple yet efficient reusing and refining training strategy, termed self-transfer training. We demonstrated this strategy would improve the model performance. We modified the SFCN (Simple Fully Convolutional Network) model to classify the working memory capacity using structural MRI data. In this study, we analyzed 5469 subjects, of which 3501 have low working memory capacity, and 1968 have high-memory capacity. First, we fully trained the model using distinct sets of hyperparameters and obtained a mean balanced accuracy of 67.80%. Further training the model using the Self-Transfer Training strategy improved the model performance to 72.53%. We conclude that i) the brain imaging features presented in the shallow layers of the models are more generic and can be trained initially and reused, while brain imaging features in later layers are more task-specific and can be refined via further training, and ii) the self-transfer training will aid in improving the model’s performance with limited samples to learn.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Mining World Indicators for Analyzing and Modeling the Development of Countries

引用

ACM/IMS Transactions on data science 2021年第4期2卷 1–27页

作者： Huang, Hong Chi, Mingyuan Song, Yu Jin, Hai The National Engineering Research Center for Big Data Technology and System Key Laboratory of Service Computing Technology and System Ministry of Education School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China

The world indicators released by the World Bank or other organizations usually give the basic public knowledge about the world. However, separate and static index lacks the complex interplay among different indicators and thus cannot help us have an overall understanding of the world. To this end, we study the world indicators from a different angle. Firstly, we discover that there exist correlations between indicators either from a static view or from a dynamic view. Moreover, taking the trade and diplomatic relationships into consideration, we construct a multi-relational network to depict the interactions between different countries, and propose a Multiple Relations to Vector (MR2vec) model to study world indicators from a network perspective. The experimental results show the changes of world indicators are predictable with the proposed model, and our proposed MR2vec has wide adaptability in predicting multi-relation networks. © 2022 Copyright held by the owner/author(s).

关键词： data mining

来源：评论

学校读者我要写书评

暂无评论

An Exploratory Computational Approach for Elucidating a Nonstructural Protein Interaction in Dengue Virus 2

引用

Procedia computer science 2024年 245卷 470-479页

作者： Alam Ahmad Hidayat Rudi Nirwantono Bens Pardamean Mathematics Department School of Computer Science Bina Nusantara University Indonesia 11480 Biotechnology Department Faculty of Engineering Bina Nusantara University Indonesia 11480 Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta Indonesia 11480 Computer Science Department BINUS Graduate Program Bina Nusantara University Jakarta Indonesia 11480

As the dengue infection still impacts hundreds of millions of people globally, unprecedented efforts in dengue drug development have been more progressive in recent decades. Computational methods provide a fast, sustainable, and efficient screening of active compounds and newly created drug molecules, including those specifically targeting nonstructural proteins (NS) of dengue viruses. In this work, protein modeling for the NS proteins of DENV-2/16681 strain was performed using a template-based homology modeling for the NS3 protein and an Artificial Intelligence (ai)-based prediction via AlphaFold for the NS4B protein. Moreover, the protein-protein interaction between the two structures was predicted using the HADDOCK server, which employs information about active and passive residues of the interaction interface to guide the docking process. After the modeling and its respective refinement process, the predicted structures of NS3 and NS4B improved their steric clashing scoring from MolProbity assessment. The refined models were then docked, and the resulting docking pose was analyzed to extract the interacting residues based on the polar contacts within the interface of the two proteins. Our result presents a preliminary study to create a dataset related to in silico molecular interactions of the NS3-NS4B interaction of different DENV types. It is helpful for building a computational pipeline for elucidating protein-ligand problems in dengue drug screenings.

关键词： bioinformatics computational biology dengue protein docking structural biology

来源：评论

学校读者我要写书评

暂无评论

Contactless Respiratory Rate Monitoring For ICU Patients Based On Unsupervised Learning

Contactless Respiratory Rate Monitoring For ICU Patients Bas...

引用

2023 IEEE/CVF Conference on computer Vision and Pattern Recognition Workshops, CVPRW 2023

作者： Liu, Zimeng Huang, Bin Lin, Chun-Liang Wu, Chieh-Liang Zhao, Changchen Chao, Wen-Cheng Wu, Yu-Cheng Zheng, Yadan Wang, Zhiru Beihang University Hangzhou Innovataion Institute China Beihang University School of Automation Science and Electrical Engineering China National Chung Hsing University College of Electrical Engineering and Computer Science Taiwan Taichung Veterans General Hospital Department of Critical Care Medicine Taiwan National Chung Hsing University College of Medicine Department of Post-Baccalaureate Medicine Taiwan National Chung Hsing University Big Data Center Taiwan Beijing University of Posts and Telecommunications Computer Science School China Beijing Foreign Studies University Graduate School of Translation and Interpretation China

ISBN: (纸本)9798350302493

Recently, the task of contactless physiological signal monitoring based on deep learning technologies has attracted a large number of scholars. However, few studies focus on the application of real-world scenarios, especially in clinical medicine scenes. In this paper, a novel video-based contactless respiratory rate measurement algorithm is developed for the Intensive Care Unit (ICU) patients. Firstly, a large-scale clinical real-world database towards ICU patient is collected in this study. Then, based on the dataset, the unsupervised learning is first introduced to extract the respiration waveform from the chest area of patients. Lastly, a respiratory rate estimator based on neural networks is proposed and trained on a periodical physiological signal simulation dataset, and utilizes the transfer learning technique to extract the respiratory rate from only 10-second respiration waveform. We obtained estimated respiratory rate with an MAE of 2.8 breaths/min and an STD of 3.0 breaths/min against the reference value computed from the specialized medical device. Extensive experiments demonstrate that our proposed methods achieve competitive results over state-of-the-art (SOTA) method in the real-world scenario. © 2023 IEEE.

关键词： Intensive care units

来源：评论

学校读者我要写书评

暂无评论

XoT (Xducation of Things): Harnessing ai and edge computing to educate all things

引用

IEEE Access 2024年

作者： Nurtantyana, Rio Hwang, Wu-Yuin Hariyanti, Uun Research Center for Data and Information Sciences Bandung Indonesia Telkom University School of Computing Indonesia Telkom University Center of Excellence Artificial Intelligence for Learning and Optimization Bandung Indonesia National Dong Hwa University College of Science and Engineering Department of Computer Science and Information Engineering Taiwan National Central University Graduate Institute of Network Learning Technology Taiwan Universitas Brawijaya Faculty of Computer Science Department of Information System Indonesia

Most English foreign language (EFL) studies focus solely on human beings. This research explores how edge computing can facilitate learning for all things. The XoT (Xducation of Things) framework was proposed to educate both human and all things. All things encompass two terms: ai-Agent and smartthings (covering physical and digital smart objects). At the core of this framework is Smart Question Answer Forwarding Mechanism (SQA-Forwarding), specifically designed to assist all things in building knowledge. To demonstrate this, the smartXoT environment was developed based on XoT framework, and its impacts on EFL learners was assessed. A quasi-experimental study involving 26 EFL learners, divided into an experimental group (EG) and a control group (CG), examined the differences in learning achievement of smartthings and EFL learners when using the smartXoT environment with/without SQA-Forwarding. Findings, on one hand, indicated that smartthings in the EG developed knowledge bases greater than those in the CG. On the other hand, the interaction between EFL learners and smartthings with SQA-Forwarding significantly improved learners' writing skills, with revisions playing a crucial role in enhancing writing quality. Thus, the XoT framework offers a novel and promising approach to educating both humans and all things. © 2013 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

On Pipelined GCN with Communication-Efficient Sampling and Inclusion-Aware Caching

On Pipelined GCN with Communication-Efficient Sampling and I...

引用

IEEE Annual Joint Conference: INFOCOM, IEEE computer and Communications Societies

作者： Shulin Wang Qiang Yu Xiong Wang Yuqing Li Hai Jin National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China School of Cyber Science and Engineering Wuhan University Wuhan China

ISBN: (数字)9798350383508

ISBN: (纸本)9798350383515

Graph convolutional network (GCN) has achieved enormous success in learning structural information from unstructured data. As graphs become increasingly large, distributed training for GCNs is severely prolonged by frequent cross-worker communications. Existing efforts to improve the training efficiency often come at the expense of GCN performance, while the communication overhead persists. In this paper, we propose PSC-GCN, a holistic pipelined framework for distributed GCN training with communication-efficient sampling and inclusion-aware caching, to address the communication bottleneck while ensuring satisfactory model performance. Specifically, we devise an asynchronous pre-fetching scheme to retrieve stale statistics (features, embedding, gradient) of boundary nodes in advance, such that the embedding aggregation and model update are pipelined with statistics transmission. To alleviate communication volume and staleness effect, we introduce a variance-reduction based sampling policy, which prioritizes inner nodes over boundary ones for reducing the access frequency to remote neighbors, thus mitigating cross-worker statistics exchange. Complementing graph sampling, a feature caching module is co-designed to buffer hot nodes with high inclusion probability, ensuring that frequently sampled nodes will be available in local memory. Extensive evaluations on real-world datasets show the superiority of PSC-GCN over state-of-the-art methods, where we can reduce training time by 72%-80% without sacrificing model accuracy.

关键词： Training Estimation error Accuracy Graph convolutional networks Computational modeling Probability Benchmark testing

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：