检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Zou, Yingtian Kawaguchi, Kenji Liu, Yingnan Liu, Jiashuo Lee, Mong-Li Hsu, Wynne School of Computing National University of Singapore Singapore Institute of Data Science National University of Singapore Singapore Department of Computer Science & Technology Tsinghua University China

Generalizing to out-of-distribution (OOD) data or unseen domain, termed OOD generalization, still lacks appropriate theoretical guarantees. Canonical OOD bounds focus on different distance measurements between source and target domains but fail to consider the optimization property of the learned model. As empirically shown in recent work, the sharpness of learned minima influences OOD generalization. To bridge this gap between optimization and OOD generalization, we study the effect of sharpness on how a model tolerates data change in domain shift which is usually captured by "robustness" in generalization. In this paper, we give a rigorous connection between sharpness and robustness, which gives better OOD guarantees for robust algorithms. It also provides a theoretical backing for "flat minima leads to better OOD generalization". Overall, we propose a sharpness-based OOD generalization bound by taking robustness into consideration, resulting in a tighter bound than non-robust guarantees. Our findings are supported by the experiments on a ridge regression model, as well as the experiments on deep learning classification tasks. Copyright © 2024, The Authors. All rights reserved.

关键词： Regression analysis

来源：评论

学校读者我要写书评

暂无评论

Generative AI: A Review on Models and Applications 2

Generative AI: A Review on Models and Applications

引用

2nd IEEE International Conference on Communication, Security and Artificial Intelligence, ICCSAI 2023

作者： Kaswan, Kuldeep Singh Dhatterwal, Jagjit Singh Malik, Kiran Baliyan, Anupam School of Computing Science and Engineering Galgotias University UP Greater Noida India Koneru Lakshmaiah Education Foundation Artificial Intelligence and Data Science AP Guntur India Matu Ram Institute of Engineering and Management Computer Science and Engineering Haryana Rohtak India Chandigarh University Department of Computer Science and Engineering Chandigarh India

ISBN: (纸本)9798350369960

Generative Artificial Intelligence (AI) stands as a transformative paradigm in machine learning, enabling the creation of complex and realistic data from latent representations. This review paper comprehensively surveys the landscape of Generative AI, encompassing its foundational concepts, diverse models, training methodologies, applications, challenges, recent advancements, evaluation metrics, and ethical dimensions. The paper begins by introducing Generative AI's significance across various domains, presenting its pivotal role in producing synthetic data with applications spanning image synthesis, text generation, music composition, drug discovery, and more. The objectives lie in elucidating the foundational concepts, delving into model intricacies, unveiling the training procedures, exploring its application landscape, addressing challenges, envisioning future directions, and discussing ethical ramifications. The foundational section elucidates the diverse array of generative models, including Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), flow-based models, Generative Reinforcement Learning (GRL), and advanced hybrid architectures. Subsequently, evaluation metrics ranging from Inception Score to perceptual similarity metrics and human evaluations are surveyed to assess generative model performance. Finally, ethical considerations underscore the necessity for addressing biases, misuse, intellectual property concerns, and the call for responsible AI development and regulation in the Generative AI landscape. © 2023 IEEE.

关键词： Advanced Hybrid Architectures Flow-Based Models Generative Adversarial Networks (GANs) Generative Reinforcement Learning (GRL) Variational Autoencoders (VAEs)

来源：评论

学校读者我要写书评

暂无评论

Mobile-enabled Prosthetic System with Machine Learning Support

Mobile-enabled Prosthetic System with Machine Learning Suppo...

引用

2024 IEEE International Conference on E-Health Networking, Application and Services, HealthCom 2024

作者： Williams, Christian Lee, Andrew Tolah, Ahmed Einafshar, Mohammadjavad Massaad, Elie Kiapour, Ali Yu, Chen-Hsiang School of Computing and Data Science Wentworth Institute of Technology BostonMA United States Department of Biomedical Engineering Wentworth Institute of Technology BostonMA United States Department of Materials and Production Aalborg University Aalborg Denmark Massachusetts General Hospital Harvard Medical School BostonMA United States College of Engineering Northeastern University BostonMA United States

ISBN: (纸本)9798350350548

Machine learning (ML) and its applications have expanded over the past few years. ML inevitably makes its way to the world of prosthetics and amputees. Smart prosthetics have been studied and growing recently. Through the collection of data from these devices, results can help the user in numerous ways. On the other hand, mobile devices and applications are widely used. However, how to combine mobile applications and ML to enhance the prosthetic device was less addressed. In this research, we studied ML with an IoT-based prosthetic device paired with a mobile application. The hypothesis was that the ML methods could help evaluate the user’s status. The study results showed that some evaluated ML methods were able to see through the average temperature, humidity and contraction percentage of the people who wear a designed prosthetic device. The results also indicated that the users could tell if the contractions reached a concerning level. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

HiSMatch: Historical Structure Matching based Temporal Knowledge Graph Reasoning

HiSMatch: Historical Structure Matching based Temporal Knowl...

引用

2022 Findings of the Association for Computational Linguistics: EMNLP 2022

作者： Li, Zixuan Hou, Zhongni Guan, Saiping Jin, Xiaolong Peng, Weihua Bai, Long Lyu, Yajuan Li, Wei Guo, Jiafeng Cheng, Xueqi School of Computer Science and Technology University of Chinese Academy of Sciences China CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences China Baidu Inc

A Temporal Knowledge Graph (TKG) is a sequence of KGs with respective timestamps, which adopts quadruples in the form of (subject, relation, object, timestamp) to describe dynamic facts. TKG reasoning has facilitated many real-world applications via answering such queries as (query entity, query relation, ?, future timestamp) about future. This is actually a matching task between a query and candidate entities based on their historical structures, which reflect behavioral trends of the entities at different timestamps. In addition, recent KGs provide background knowledge of all the entities, which is also helpful for the matching. Thus, in this paper, we propose the Historical Structure Matching (HiSMatch) model. It applies two structure encoders to capture the semantic information contained in the historical structures of the query and candidate entities. Besides, it adopts another encoder to integrate the background knowledge into the model. TKG reasoning experiments on six benchmark datasets demonstrate the significant improvement of the proposed HiSMatch model, with up to 5.6% performance improvement in MRR, compared to the state-of-the-art baselines. © 2022 Association for Computational Linguistics.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

TCM Tongue Segmentation and Analysis with a Mobile Device 6

TCM Tongue Segmentation and Analysis with a Mobile Device

引用

6th IEEE International Conference on Electrical, Electronics and System Engineering, ICEESE 2023

作者： Grant, Alex Ferrarelli, James Rozano, Maximilian Liu, Qingqing Su, Jiaming Yu, Chen-Hsiang School of Computing and Data Science Wentworth Institute of Technology BostonMA United States Dongzhimen Hospital Beijing University of Chinese Medicine Beijing China Multidisciplinary Graduate Engineering College of Engineering Northeastern University BostonMA United States

ISBN: (纸本)9781665453714

Mobile devices and applications provide increased accessibility to things like Traditional Chinese Medicine (TCM). Ideally, one would be able to seek out a diagnosis to the tongue using advanced machine learning models in a mobile device. However, such kind of models are still not yet available for the public. In this research, we investigated a way of combining machine learning with a mobile application to support tongue diagnosis for the public. For tongue segmentation, we experimented with an unsupervised approach (K-means), a supervised method and a deep learning model. There are two outcomes presented in this research. Firstly, the user can seamlessly take a picture from a mobile phone and send it over to the server for a segmentation process. Secondly, the image processed by the server would have at least some levels of autonomous analysis preformed on it with acceptable results using machine learning methods. Our current results indicate that we were able to finish some key steps in this process. © 2023 IEEE.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

Maxflow-Based Bounds for Low-Rate Information Propagation over Noisy Networks

arXiv

引用

arXiv 2023年

作者： Ling, Yan Hao Scarlett, Jonathan Department of Computer Science Department of Mathematics School of Computing National University of Singapore Singapore Institute of Data Science United States

We study error exponents for the problem of low-rate communication over a directed graph, where each edge in the graph represents a noisy communication channel, and there is a single source and destination. We derive maxflow-based achievability and converse bounds on the error exponent that match when there are two messages and all channels satisfy a symmetry condition called pairwise reversibility. More generally, we show that the upper and lower bounds match to within a factor of 4. We also show that with three messages there are cases where the maxflow-based error exponent is strictly suboptimal, thus showing that our tightness result cannot be extended beyond two messages without further assumptions. Copyright © 2023, The Authors. All rights reserved.

关键词： Directed graphs

来源：评论

学校读者我要写书评

暂无评论

DOMAIN ADVERSARIAL TRAINING ON CONDITIONAL VARIATIONAL AUTO-ENCODER FOR CONTROLLABLE MUSIC GENERATION 23

DOMAIN ADVERSARIAL TRAINING ON CONDITIONAL VARIATIONAL AUTO-...

引用

23rd International Society for Music Information Retrieval Conference, ISMIR 2022

作者： Zhao, Jingwei Xia, Gus Wang, Ye School of Computing NUS Singapore Institute of Data Science NUS Singapore Music X Lab. NYU Shanghai China Integrative Sciences and Engineering Programme NUS Graduate School Singapore MBZUAI

ISBN: (纸本)9781732729926

The variational auto-encoder has become a leading framework for symbolic music generation, and a popular research direction is to study how to effectively control the generation process. A straightforward way is to control a model using different conditions during inference. However, in music practice, conditions are usually sequential (rather than simple categorical labels), involving rich information that overlaps with the learned representation. Consequently, the decoder gets confused about whether to ªlisten toºthe latent representation or the condition, and sometimes just ignores the condition. To solve this problem, we leverage domain adversarial training to disentangle the representation from condition cues for better control. Specifically, we propose a condition corruption objective that uses the representation to denoise a corrupted condition. Minimized by a discriminator and maximized by the VAE encoder, this objective adversarially induces a condition-invariant representation. In this paper, we focus on the task of melody harmonization1 to illustrate our idea, while our methodology can be generalized to other controllable generative tasks. Demos and experiments show that our methodology facilitates not only condition-invariant representation learning but also higher-quality controllability compared to baselines. © J. Zhao, G. Xia, and Y. Wang.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

A Survey on Cyber Attack Detection: Techniques, datasets, and Challenge

A Survey on Cyber Attack Detection: Techniques, Datasets, an...

引用

Research Methodologies in Knowledge Management, Artificial Intelligence and Telecommunication Engineering (RMKMATE), International Conference on

作者： G Geetha Manjula Rajagopal K Purna Chand Department of Computing Technologies School of Computing (Research Scholor) SRM Institute of Science and Technology Kattankulathur Tamilnadu Department of Computing Technologies School of Computing (Assistant Professor) SRM Institute of Science and Technology Kattankulathur Tamilnadu Department of Computer Science and Engineering(Data Science) (Professor) B V Raju Institute of Technology Narsapur India

Networking technology is expanded along with cyber threats. Cyber threats are the instances which adversely influences the functional operations of the system or network. It results in various issues such as illegal access, data leakage, data stolen, DDoS attacks, etc. To overcome such issues, several intrusion detection systems are implemented over the past years. Though, these intrusion detection systems exhibit several limitations while trying to achieve excellent efficiency in attack detection. In this survey, totally 50 cyber-attack detection based research works are analysed under several perspectives. The analysis is done on several IDS techniques implemented in existing works. The reviewed learning-based approaches are explained in two major categories as ML and DL approaches. Here, the ML algorithms are further classified into three types: supervised, unsupervised and reinforcement learning approaches. Moreover, the existing works are also analysed in accordance with optimization algorithms and it is classified as metaheuristic and stochastic algorithms. In which, population-based algorithms and swarm-based algorithms are the categories classified under metaheuristic algorithms. In addition, the analyses are also performed on datasets usage and cyber-attack mitigation models. Furthermore, the analysis on existing works is done on performance analysis. And it is analysed under performance of cyber-attack detection models with specific description about maximum and minimum performance attained by recent works on cyber-attack detection models. Finally, the chronological review and challenges involved in existing cyber-attack detection models are discussed.

关键词：

来源：评论

学校读者我要写书评

暂无评论

BEAT TRANSFORMER: DEMIXED BEAT AND DOWNBEAT TRACKING WITH DILATED SELF-ATTENTION 23

BEAT TRANSFORMER: DEMIXED BEAT AND DOWNBEAT TRACKING WITH DI...

引用

23rd International Society for Music Information Retrieval Conference, ISMIR 2022

ISBN: (纸本)9781732729926

We propose Beat Transformer, a novel Transformer encoder architecture for joint beat and downbeat tracking. Different from previous models that track beats solely based on the spectrogram of an audio mixture, our model deals with demixed spectrograms with multiple instrument channels. This is inspired by the fact that humans perceive metrical structures from richer musical contexts, such as chord progression and instrumentation. To this end, we develop a Transformer model with both time-wise attention and instrument-wise attention to capture deep-buried metrical cues. Moreover, our model adopts a novel dilated self-attention mechanism, which achieves powerful hierarchical modelling with only linear complexity. Experiments demonstrate a significant improvement in demixed beat tracking over the non-demixed version. Also, Beat Transformer achieves up to 4% point improvement in downbeat tracking accuracy over the TCN architectures. We further discover an interpretable attention pattern that mirrors our understanding of hierarchical metrical structures. © J. Zhao, G. Xia, and Y. Wang.

关键词： Spectrographs

来源：评论

学校读者我要写书评

暂无评论

Deep Spectral Copula Mechanisms Modeling Coupled and Volatile Multivariate Time Series

Deep Spectral Copula Mechanisms Modeling Coupled and Volatil...

引用

International Conference on data science and Advanced Analytics (DSAA)

作者： Yang Yang Zhilin Zhao Longbing Cao Data Science Institute University of Technology Sydney Sydney Australia School of Computing Macquarie University Sydney Australia

Exploring inter- and intra-time series relations and handling volatile covariates form various challenges in modeling Coupled and Volatile Multivariate Time Series (CVMTS). A typical CVMTS data is the COVID-19 case time series across multiple countries, whose covariates may involve high volatility caused by missing samples. The existing approaches merely focus on a single set of multivariate time series or multiple multivariate time series without considering their volatile temporal covariates. They do not sufficiently characterize CVMTS features by explicitly modeling intra- and inter-MTS couplings and effectively handling volatile covariates in multiple multivariate time series. Accordingly, we propose Deep Spectral Copula Mechanisms (DSCM) to adapt CVMTS. Specifically, DSCM (1) incorporates a Singular Spectral Analysis (SSA) module to reduce the volatility of multiple covariates; (2) applies an intra-MTS coupling module to explicitly model the temporal couplings within a single set of multivariate time series; and (3) transforms target variables into joint probability distributions by Gaussian copula transformation to establish inter-MTS couplings across multiple multivariate time series. Substantial experiments on COVID-19 time-series data from multiple countries indicate the superiority of DSCM over state-of-the-art approaches.

关键词：

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：