检索结果-内蒙古大学图书馆

27th World Wide web (WWW) Conference

作者： Wei, Hong Zhou, Hao Sankaranarayanan, Jangan Sengupta, Sudipta Samet, Hanan Univ Maryland Dept Comp Sci College Pk MD 20742 USA Univ Maryland College Pk MD 20742 USA Amazon Web Serv AWS Seattle WA USA

ISBN: (纸本)9781450356404

The tweet count prediction of a local spatial region is to forecast the number of tweets that are likely to be posted from that area over a relatively short period of time. It has many applications such as human mobility analysis, traffic planning, and abnormal event detection. In this paper, we formulate tweet count prediction as a spatiotemporal sequence forecasting problem and design an end-to-end convolutional LSTM based network with skip connection for this problem. Such a model enables us to exploit the unique properties of spatiotemporal data, consisting of not only the temporal characteristics such as temporal closeness, period and trend properties, but also spatial dependencies. Our experiments on the city of Seattle, WA as well as a larger city of New York City show that the proposed method consistently outperforms the competitive baseline approaches.

关键词： Social Network Twitter Tweet Count Prediction LSTM Convolution Convolutional LSTM Residual Neural Network

来源：评论

学校读者我要写书评

暂无评论

AI model disgorgement: Methods and choices

引用

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA 2024年第18期121卷 e2307304121页

作者： Achille, Alessandro Kearns, Michael Klingenberg, Carson Soatto, Stefano Amazon Web Serv Artificial Intelligence AWS AI Pasadena CA 91125 USA Univ Penn Comp & Informat Sci Philadelphia PA 19130 USA Univ Calif Los Angeles Comp Sci Los Angeles CA 90095 USA

Over the past few years, machine learning models have significantly increased in size and complexity, especially in the area of generative AI such as large language models. These models require massive amounts of data and compute capacity to train, to the extent that concerns over the training data (such as protected or private content) cannot be practically addressed by retraining the model "from scratch" with the questionable data removed or altered. Furthermore, despite significant efforts and controls dedicated to ensuring that training corpora are properly curated and composed, the sheer volume required makes it infeasible to manually inspect each datum comprising a training corpus. One potential approach to training corpus data defects is model disgorgement, by which we broadly mean the elimination or reduction of not only any improperly used data, but also the effects of improperly used data on any component of an ML model. Model disgorgement techniques can be used to address a wide range of issues, such as reducing bias or toxicity, increasing fidelity, and ensuring responsible use of intellectual property. In this paper, we survey the landscape of model disgorgement methods and introduce a taxonomy of disgorgement techniques that are applicable to modern ML systems. In particular, we investigate the various meanings of "removing the effects" of data on the trained model in a way that does not require retraining from scratch.

关键词： machine learning artificial intelligence model disgorgement machine unlearning generative AI

来源：评论

学校读者我要写书评

暂无评论

Containerized deep learning in agriculture: Orchestrating GoogleNet with Kubernetes on high performance computing

引用

CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE 2024年第15期36卷

作者： Hasan, Syed Humaid Hasan, Syeda Huyam Khan, Usman Ali Hasan, Syed Hamid Amazon Web Serv AWS Deployment Containers Dallas TX USA Amazon Web Serv AWS Database Dallas TX USA King Abdulaziz Univ Coll Comp Sci & Informat Technol Dept Informat Syst Jeddah Saudi Arabia

Smart Farming has become a cornerstone of modern agriculture, offering data-driven insights and automation that optimize resource utilization and increase crop yields. The use of cutting-edge technologies in agriculture has given rise to Smart Farming, which has transformed traditional farming practices into efficient, data-driven operations. This paper explores the synergy between high-performance computing (HPC) systems, Kubernetes orchestration, GoogleNet architecture, and containerization to redefine the future of farming. At the heart of this transformation lies the GoogleNet architecture, a deep learning powerhouse recognized for its efficiency and accuracy in image recognition tasks. The orchestration capabilities of Kubernetes, a versatile tool for managing containerized workloads efficiently on HPC clusters. Hence, in this work, we investigate the intricacies of deploying GoogleNet-based deep learning models within containerized environments orchestrated by Kubernetes on HPC infrastructure. It explores resource optimization, scalability, security, and adaptability, all tailored to the unique demands of the agricultural domain to evaluate the effectiveness of the given technique it is compared with the existing techniques namely Hermes, Horus, CYBELE, and RZ-SHAN. The attained ranges of proposed method of various measures of accuracy, precision, recall, and F1-score are 98.65%, 97.45%, 97.87%, and 98.12% for the Pilot Wheat Ear dataset. Also, the processing time for the proposed approach is 181.50 and 120.2 m for the Pilot Wheat Ear Dataset and the Pilot soya bean farming dataset. The latency of the proposed approach attains a lower value of 1.5 and 1.1 s pilot soya bean farming dataset and Pilot Wheat Ear dataset. The experimental outcome demonstrates the efficiency of the proposed approaches to improve Smart Farming agriculture.

关键词： agriculture GoogleNet high performance Kubernetes orchestration smart farming wingsuit

来源：评论

学校读者我要写书评

暂无评论

Optimized Bi-LSTM: a novel approach for attack detection in industrial IoT

引用

SIGNAL IMAGE AND VIDEO PROCESSING 2024年第5期18卷 4903-4913页

作者： Alassafi, Madini O. Hasan, Syed Humaid Badri, Sahar Hasan, Syed Hamid King Abdulaziz Univ Coll Comp Sci & Informat Technol Dept Informat Technol Jeddah Saudi Arabia Amazon Web Serv AWS Dallas TX USA King Abdulaziz Univ Coll Comp Sci & Informat Technol Dept Informat Syst Jeddah Saudi Arabia

The Industrial Internet of Things (IIoT) plays an important role as its implementation requires scalable and secure methods. IIoT integrates databases, machines, various sensors, and services to improve our lives in many ways, including e-healthcare and smarter cities. Accurately identifying attacks in IIoT is critical to mitigating security risks, and the volume of such attacks is rising to an unprecedented level. As IIoT solutions integrate into larger operational systems, additional planning and development of a new mechanism is required to ensure privacy and security. Many deep learning approaches have been developed to focus on intrusion detection. However, such approaches suffer from low detection accuracy, efficiency, privacy, and high time consumption. To mitigate these issues, this study proposes a novel approach called bi-long short-term memory (Bi-LSTM) in IIoT security for privacy and security in industrial systems. To improve the performance of Bi-LSTM, we use modified coot optimization. Our proposed Bi-LSTM approach effectively detects attacks in the IIoT environment. To assess its performance, we analyze various evaluation measures including precision, sensitivity, Matthews correlation coefficient (MCC), and accuracy. The results of our performance analysis reveal that the proposed optimized Bi-LSTM method achieves a higher performance rate of 98.91% accuracy, 95.86% sensitivity, 95.63% MCC, and 97.52% accuracy than other existing approaches. The existing approaches such as CDW_FedAvg, RNN, COSNN, and RCS-DOFH achieved lower accuracy of 98.52%, 97.83%, 95.17%, and 96.24%, respectively. The findings highlight the potential for improved cybersecurity in the IIoT, emphasizing the importance of effective planning and innovative critical systems.

关键词： Industrial internet of things Cybersecurity Bi-long short-term memory (Bi-LSTM) Coot optimization Attack detection Cloud computing Blockchain

来源：评论

学校读者我要写书评

暂无评论

Deploying warehouse robots with confidence: the BRAIN-IoT framework's functional assurance

引用

JOURNAL OF SUPERCOMPUTING 2024年第1期80卷 1206-1237页

作者： Baouya, Abdelhakim Chehida, Salim Bensalem, Saddek Gurgen, Levent Nicholson, Richard Cantero, Miquel Diaznava, Mario Ferrera, Enrico Univ Toulouse CNRS UT2 IRIT Toulouse France Univ Grenoble Alpes VERIMAG Grenoble France Kentyou Grenoble France Amazon Web Serv AWS London England Robotnik Automat Valencia Spain STMicroelectronics Grenoble France LINKS Fdn Turin Italy

Our study details the development and validation of an orchestrator-controlled robotic network that effectively organizes and manages the activities of multiple robots. The design workflow is based on a model-driven methodology that allows for the independent specification of robot behaviour, which can be successfully refined regardless of the physical architecture. The main focus of this study involves the verification and analysis of robot orchestration by building formal models in a component-port- connector fashion supported by BIP language (behaviour-interaction-priority). The model also helps to study the automated orchestration with the help of a centralized computer tasks manager. The related functional requirements gathered from industrial partners are specified in temporal logic. Statistical model checking is performed to verify the model's correctness, providing a functional assurance to achieve the deployment. Validation is a carry out using a dedicated robotic platform simulator. We demonstrate the capability of the verification artefact for the Brain-IoT (https:// cordis. ***/project/id/780089) platform and ways of applying them to potentially com-plex case studies.

关键词： Internet of Things Statistical model checking Robotics Scalability

来源：评论

学校读者我要写书评

暂无评论

Variance Reduced Training with Stratified Sampling for Forecasting Models 38

Variance Reduced Training with Stratified Sampling for Forec...

引用

International Conference on Machine Learning (ICML)

作者： Lu, Yucheng Park, Youngsuk Chen, Lifan Wang, Yuyang De Sa, Christopher Foster, Dean Cornell Univ Dept Comp Sci Ithaca NY 14853 USA Amazon Web Serv AWS AI Labs Palo Alto CA 94301 USA Amazon Res New York NY USA Univ Penn Philadelphia PA 19104 USA

ISBN: (纸本)9781713845065

In large-scale time series forecasting, one often encounters the situation where the temporal patterns of time series, while drifting over time, differ from one another in the same dataset. In this paper, we provably show under such heterogeneity, training a forecasting model with commonly used stochastic optimizers (e.g. SGD) potentially suffers large variance on gradient estimation, and thus incurs long-time training. We show that this issue can be efficiently alleviated via stratification, which allows the optimizer to sample from pre-grouped time series strata. For better trading-off gradient variance and computation complexity, we further propose SCott (Stochastic Stratified Control Variate Gradient Descent), a variance reduced SGD-style optimizer that utilizes stratified sampling via control variate. In theory, we provide the convergence guarantee of SCott on smooth non-convex objectives. Empirically, we evaluate SCott and other baseline optimizers on both synthetic and real-world time series forecasting problems, and demonstrate SCott converges faster with respect to both iterations and wall clock time.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

AoI, Timely-Throughput, and Beyond: A Theory of Second-Order Wireless Network Optimization

引用

IEEE-ACM TRANSACTIONS ON NETWORKING 2024年第6期32卷 4707-4721页

作者： Guo, Daojing Nakhleh, Khaled Hou, I-Hong Kompella, Sastry Kam, Clement Texas A&M Univ ECE Dept College Stn TX 77843 USA Amazon Web Serv AWS Seattle WA 98109 USA Nexcepta Inc Gaithersburg MD 20878 USA US Naval Res Lab NRL Wireless Network Theory Sect Washington DC 20375 USA

This paper introduces a new theoretical framework for optimizing second-order behaviors of wireless networks. Unlike existing techniques for network utility maximization, which only consider first-order statistics, this framework models every random process by its mean and temporal variance. The inclusion of temporal variance makes this framework well-suited for modeling Markovian fading wireless channels and emerging network performance metrics such as age-of-information (AoI) and timely-throughput. Using this framework, we sharply characterize the second-order capacity region of wireless access networks. We also propose a simple scheduling policy and prove that it can achieve every interior point in the second-order capacity region. To demonstrate the utility of this framework, we apply it to an unsolved network optimization problem where some clients wish to minimize AoI while others wish to maximize timely-throughput. We show that this framework accurately characterizes AoI and timely-throughput. Moreover, it leads to a tractable scheduling policy that outperforms other existing work.

关键词： Age-of-information (AoI) timely-throughput Brownian motion wireless networks

来源：评论

学校读者我要写书评

暂无评论

Transformers from an Optimization Perspective 36

Transformers from an Optimization Perspective

引用

36th Conference on Neural Information Processing Systems (NeurIPS)

作者： Yang, Yongyi Huang, Zengfeng Wipf, David Univ Michigan Ann Arbor MI 48109 USA Fudan Univ Shanghai Peoples R China Amazon Web Serv Seattle WA USA AWS Shanghai AI Lab Shanghai Peoples R China

ISBN: (纸本)9781713871088

Deep learning models such as the Transformer are often constructed by heuristics and experience. To provide a complementary foundation, in this work we study the following problem: Is it possible to find an energy function underlying the Transformer model, such that descent steps along this energy correspond with the Transformer forward pass? By finding such a function, we can view Transformers as the unfolding of an interpretable optimization process across iterations. This unfolding perspective has been frequently adopted in the past to elucidate more straightforward deep models such as MLPs and CNNs;however, it has thus far remained elusive obtaining a similar equivalence for more complex models with self-attention mechanisms like the Transformer. To this end, we first outline several major obstacles before providing companion techniques to at least partially address them, demonstrating for the first time a close association between energy function minimization and deep layers with self-attention. This interpretation contributes to our intuition and understanding of Transformers, while potentially laying the ground-work for new model designs.

关键词： Optimization

来源：评论

学校读者我要写书评

暂无评论

Correcting Exposure Bias for Link Recommendation 38

Correcting Exposure Bias for Link Recommendation

引用

International Conference on Machine Learning (ICML)

作者： Gupta, Shantanu Wang, Hao Lipton, Zachary Wang, Yuyang Amazon Seattle WA USA Carnegie Mellon Univ Machine Learning Dept Pittsburgh PA 15213 USA Rutgers State Univ Dept Comp Sci New Brunswick NJ USA Amazon Web Serv AWS AI Labs Palo Alto CA USA

ISBN: (纸本)9781713845065

Link prediction methods are frequently applied in recommender systems, e.g., to suggest citations for academic papers or friends in social networks. However, exposure bias can arise when users are systematically underexposed to certain relevant items. For example, in citation networks, authors might be more likely to encounter papers from their own field and thus cite them preferentially. This bias can propagate through naively trained link predictors, leading to both biased evaluation and high generalization error (as assessed by true relevance). Moreover, this bias can be exacerbated by feedback loops. We propose estimators that leverage known exposure probabilities to mitigate this bias and consequent feedback loops. Next, we provide a loss function for learning the exposure probabilities from data. Finally, experiments on semi-synthetic data based on real-world citation networks, show that our methods reliably identify (truly) relevant citations. Additionally, our methods lead to greater diversity in the recommended papers' fields of study. The code is available at ***/shantanu95/exposure-bias-link-rec.

关键词： Feedback

来源：评论

学校读者我要写书评

暂无评论

A Comparison of Four Notions of Isomorphism-Based Security for Graphs

引用

IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING 2023年第4期20卷 3371-3377页

作者： Lin, Zhiyuan Tripunitara, Mahesh Amazon Web Serv AWS Vancouver BC V6E 4G2 Canada Univ Waterloo Waterloo ON N2L 3G1 Canada Univ Waterloo Dept Elect & Comp Engn Waterloo ON N2L 3G1 Canada

A graph is a powerful abstraction for representing information. We address the problem of publishing a secure version of a graph that does not leak information to an adversary who may possess prior information about portions of the graph, and may have unbounded computational power. In this context, we revisit four notions of security, all of which are based on variants of graph isomorphism, that have been proposed in two different application contexts in the literature. We compare the four notions to one another, first from the standpoint of strength, i.e., whether meeting one notion implies meeting another, and then from the standpoint of computational hardness, i.e., what the exact computational complexity is for the problem of checking whether a graph meets a notion. For the latter, we identify that for two of the notions we consider, the problem is NP-complete, and for the two others, it is ISO-complete, where ISO is the class of problems induced by graph isomorphism. We observe that strength is not necessarily correlated to computational hardness. In summary, our work makes contributions at the foundations of an important notion of security for graphs.

关键词： Security ISO Integrated circuits Publishing Organizations Integrated circuit modeling Computational complexity

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：