检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Zhengyang Chen Bing Han Shuai Wang Yidi Jiang Yanmin Qian Department of Computer Science and Engineering Auditory Cognition and Computational Acoustics Lab MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai China Shenzhen Research Institute of Big Data Shenzhen China School of Data Science The Chinese University of Hong Kong Shenzhen China National University of Singapore Singapore

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Speaker diarization is typically considered as a discriminative task, using discriminative approaches to produce fixed diarization results. In this paper, we explore for the first time the use of neural network-based generative methods for speaker diarization. We implement a Flow-Matching (FM) based generative algorithm within the sequenceto-sequence target speaker voice activity detection (Seq2Seq-TSVAD) diarization system. Our experiments reveal that applying the generative method directly to the original binary label sequence space of the TS-VAD output is ineffective. To address this issue, we propose mapping the binary label sequence into a dense latent space before applying the generative algorithm, and our proposed Flow-TSVAD method can significantly outperform the traditional Seq2Seq-TSVAD system. Additionally, we observe that the FM algorithm converges rapidly during the inference stage, only requiring two inference steps to achieve promising results. Moreover, as a generative model, Flow-TSVAD allows for sampling different diarization results by running the model multiple times, so the ensemble system combining the results from various sampling instances can further boost the diarization performance.

关键词： Voice activity detection Frequency modulation Computational modeling Signal processing algorithms Signal processing Inference algorithms Acoustics Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Statistical, Robustness, and Computational Guarantees for Sliced Wasserstein Distances

arXiv

引用

arXiv 2022年

作者： Nietert, Sloan Sadhu, Ritwik Goldfeld, Ziv Kato, Kengo Department of Computer Science Cornell University United States Department of Statistics and Data Science Cornell University United States School of Electrical and Computer Engineering Cornell University United States

Sliced Wasserstein distances preserve properties of classic Wasserstein distances while being more scalable for computation and estimation in high dimensions. The goal of this work is to quantify this scalability from three key aspects: (i) empirical convergence rates;(ii) robustness to data contamination;and (iii) efficient computational methods. For empirical convergence, we derive fast rates with explicit dependence of constants on dimension, subject to log-concavity of the population distributions. For robustness, we characterize minimax optimal, dimension-free robust estimation risks, and show an equivalence between robust sliced 1-Wasserstein estimation and robust mean estimation. This enables lifting statistical and algorithmic guarantees available for the latter to the sliced 1-Wasserstein setting. Moving on to computational aspects, we analyze the Monte Carlo estimator for the average-sliced distance, demonstrating that larger dimension can result in faster convergence of the numerical integration error. For the max-sliced distance, we focus on a subgradient-based local optimization algorithm that is frequently used in practice, albeit without formal guarantees, and establish an O(∊−4) computational complexity bound for it. Our theory is validated by numerical experiments, which altogether provide a comprehensive quantitative account of the scalability question. Copyright © 2022, The Authors. All rights reserved.

关键词： Scalability

来源：评论

学校读者我要写书评

暂无评论

Drowsiness Detection System Using the Movement of the Eye of A Human Through Machine Learning Algorithms in Order to Avoid Accidents

Drowsiness Detection System Using the Movement of the Eye of...

引用

Advance Computing and Innovative Technologies in engineering (ICACITE), International Conference on

作者： Jenifer Jothi Mary A Subhashish Goswami B G Rahul T. Sakthika Dewank Mahajan Bhasker Pant Department Computer Science St. Joseph's College (Autonomous) Affiliated to Bharathidasan University Tiruchirappalli Department of Computer Science Dev Bhoomi Uttarakhand University Dehradun Uttarakhand India Department of Civil Engineering Koneru Lakshmaiah Educational Foundation Zoology A.P.C. Mahalaxmi College for Women Thoothukudi India Data Analyst University of Michigan Department of Computer Science & Engineering Deemed to be University Dehradun Uttarakhand India

This function is meant to keep individuals safe in the event of an accident. In this study, the development of ADAS is shown, with a focus on driver fatigue identification as the main goal of decreasing traffic fatalities. In this situation, we use artificial intelligence to anticipate the condition and sentiments of the operator, improving traffic safety. For this research, we'll use CNN, which can accurately detect driver fatigue from driving images. In this study, we can employ the Electrocardiographic to measure weariness psychologically. Artificial intelligence refers to a system's ability to learn and grow without having to be programmed. We also employ OpenCv, Keras, and Tensorflow, among other machine learning algorithms [1].Road accidents are rising quickly each year as a result of vehicle technology and mechanical developments that allow for fast driving. Road accidents claim the lives of 1.35 million people worldwide every year, with 151 thousand fatalities reported in India alone in 2016. From this, drivers are at fault for almost 78 percent of traffic incidents. Drunk driving, over speeding, and sleepiness are the main causes of these incidents, with drowsiness accounting for roughly 40% of all *** are aware of the dangers of driving after drinking, but they are not aware of the dangers of tiredness since there are no tools available to gauge driver *** a motorist is distracted while driving, it slows down their response time and affects how they steer. In order to address this issue, we'll employ machine learning to determine whether or not the driver is sleepy [2]. By setting a threshold number, we can determine if the driver is tired or not. Typically, when someone feels sleepy, the rate at which his or her eyes blink slows *** application offers a method for scanning facial features and, after recognising the face, utilising those crucial cues for eye tracking. This guarantees that the driver has complete control over his

关键词：

来源：评论

学校读者我要写书评

暂无评论

Navigating the Future of Federated Recommendation Systems with Foundation Models

arXiv

引用

arXiv 2024年

作者： Li, Zhiwei Long, Guodong Zhang, Chunxu Zhang, Honglei Jiang, Jing Zhang, Chengqi Australian Artificial Intelligence Institute Faculty of Engineering and Information Technology University of Technology Sydney UltimoNSW2007 Australia College of Computer Science and Technology Jilin University Jilin130012 China School of Computer Science and Technology Beijing Jiaotong University Beijing100044 China Department of Data Science and Artificial Intelligence The Hong Kong Polytechnic University Hong Kong

In recent years, the integration of federated learning (FL) and recommendation systems (RS), known as Federated Recommendation Systems (FRS), has attracted attention for preserving user privacy by keeping private data on client devices. However, FRSs face inherent limitations such as data heterogeneity and scarcity, due to the privacy requirements of FL and the typical data sparsity issues of RSs. Models like ChatGPT are empowered by the concept of transfer learning and self-supervised learning, so they can be easily applied to the downstream tasks after fine-tuning or prompting. These models, so-called Foundation Models (FM), focus on understanding the human’s intent and perform following their designed roles in the specific tasks, which are widely recognized for producing high-quality content in the image and language domains. Thus, the achievements of FMs inspire the design of FRS and suggest a promising research direction: integrating foundation models to address the above limitations. In this study, we conduct a comprehensive review of FRSs with FMs. Specifically, we: 1) summarise the common approaches of current FRSs and FMs;2) review the challenges posed by FRSs and FMs;3) discuss potential future research directions;and 4) introduce some common benchmarks and evaluation metrics in the FRS field. We hope that this position paper provides the necessary background and guidance to explore this interesting and emerging topic. Copyright © 2024, The Authors. All rights reserved.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

An optimized back propagation neural network for automated evaluation of health condition using sensor data

引用

Measurement: Sensors 2023年 29卷

作者： Mandala, Vishwanadham Senthilnathan, T. Suganyadevi, S. Gobhinath, S. Selvaraj, DhanaSekaran Dhanapal, R. Enterprise Architect Student in Indiana University MS in Data Science IU Bloomington 107 S. Indiana Avenue BloomingtonIN47405 United States Department of Computer Science Bengaluru_29 India Department of ECE KPR Institute of Engineering and Technology Coimbatore India Department of Electrical and Electronics Engineering Dr.N.G.P. Institute of Technology Coimbatore India Department of Electronics and Communication Engineering Sri Eshwar College of Engineering Coimbatore641202 India Department of Computer Science and Engineering Karpagam Academy of Higher Education Tamilnadu Coimbatore India

Ships and other large equipment must meet strict standards for equipment integrity and operational dependability in order to perform missions. To meet this demand, one of the essential linkages is to guarantee the long-term safe and healthy functioning of their power transmission equipment. The Optimized Back Propagation Neural Network (OBPNN) technique used in this study introduces a unique method for monitoring sensor data and evaluating the health state, with the SVM being optimized using the fish swarm algorithm (FSA). A major problem that maintenance is facing nowadays is reliable fault prediction. One of the trickiest difficulties is arguably automatically modelling typical behaviour from condition monitoring data, particularly when there is little information about actual failures. A data-driven learning framework with the best bandwidth selection is suggested to address this challenge. It is based on nonparametric density estimation for outlier identification and OBPNN for normality modelling. The distance to the separating hyper plane's log-normalization is used to provide a health score that is also available. The algorithm's viability is shown by experimental findings while evaluating the progression of a major defect over time in a marine diesel engine. Improved prediction capabilities and low false positive rates on healthy data are realized. © 2023 The Authors

关键词： Condition monitoring

来源：评论

学校读者我要写书评

暂无评论

Deep-learning-based single-domain and multidomain protein structure prediction with D-I-TASSER

引用

Nature Biotechnology 2025年 1-13页

作者： Zheng, Wei Wuyun, Qiqige Li, Yang Liu, Quancheng Zhou, Xiaogen Peng, Chunxiang Zhu, Yiheng Freddolino, Lydia Zhang, Yang NITFID School of Statistics and Data Science AAIS LPMC and KLMDASR Nankai University Tianjin China Department of Computational Medicine and Bioinformatics University of Michigan Ann ArborMI United States Department of Computer Science and Engineering Michigan State University East LansingMI United States Cancer Science Institute of Singapore National University of Singapore Singapore Singapore Department of Biological Chemistry University of Michigan Ann ArborMI United States Department of Computer Science School of Computing National University of Singapore Singapore Singapore Department of Biochemistry Yong Loo Lin School of Medicine National University of Singapore Singapore Singapore

The dominant success of deep learning techniques on protein structure prediction has challenged the necessity and usefulness of traditional force field-based folding simulations. We proposed a hybrid approach, deep-learning-based iterative threading assembly refinement (D-I-TASSER), which constructs atomic-level protein structural models by integrating multisource deep learning potentials with iterative threading fragment assembly simulations. D-I-TASSER introduces a domain splitting and assembly protocol for the automated modeling of large multidomain protein structures. Benchmark tests and the most recent critical assessment of protein structure prediction, 15 experiments demonstrate that D-I-TASSER outperforms AlphaFold2 and AlphaFold3 on both single-domain and multidomain proteins. Large-scale folding experiments further show that D-I-TASSER could fold 81% of protein domains and 73% of full-chain sequences in the human proteome with results highly complementary to recently released models by AlphaFold2. These results highlight a new avenue to integrate deep learning with classical physics-based folding simulations for high-accuracy protein structure and function predictions that are usable in genome-wide applications. © The Author(s) 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A Systematic Review of Machine Learning Approaches for Detecting Deceptive Activities on Social Media: Methods, Challenges, and Biases

arXiv

引用

arXiv 2024年

作者： Liu, Yunchong Shen, Xiaorui Zhang, Yeyubei Wang, Zhongyan Tian, Yexin Dai, Jianglai Cao, Yuchen School of Engineering and Applied Science University of Pennsylvania United States Khoury college of computer science Northeastern University United States Center for Data Science New York University United States Georgia Institute of Technology College of Computing United States Department of EECS University of California Berkeley United States

Social media platforms like Twitter, Facebook, and Instagram have facilitated the spread of misinformation, necessitating automated detection systems. This systematic review evaluates 36 studies that apply machine learning (ML) and deep learning (DL) models to detect fake news, spam, and fake accounts on social media. Using the Prediction model Risk Of Bias ASsessment Tool (PROBAST), the review identified key biases across the ML lifecycle: selection bias due to non-representative sampling, inadequate handling of class imbalance, insufficient linguistic preprocessing (e.g., negations), and inconsistent hyperparameter tuning. Although models such as Support Vector Machines (SVM), Random Forests, and Long Short-Term Memory (LSTM) networks showed strong potential, over-reliance on accuracy as an evaluation metric in imbalanced data settings was a common flaw. The review highlights the need for improved data preprocessing (e.g., resampling techniques), consistent hyperparameter tuning, and the use of appropriate metrics like precision, recall, F1 score, and AUROC. Addressing these limitations can lead to more reliable and generalizable ML/DL models for detecting deceptive content, ultimately contributing to the reduction of misinformation on social media. © 2024, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Implementing Automated Reasoning in Natural Language Processing

Implementing Automated Reasoning in Natural Language Process...

引用

International Conference on Futuristic Technologies (INCOFT)

作者： Kalyan Acharjya Rakshith N N. Thangarasu Hemalatha S Varsha D Adarsh Srivastava Maharishi School of Engineering and Technology Maharishi University of Information Technology Uttar Pradesh India Department of Computer ScienceKarpagam Academy of Higher Education Coimbatore Department of Artificial Intelligence & Data Science Vishwakarma Institute of Information Technology Pune India Department of Computer Science and Information Technology Jain (Deemed to be University) Bangalore Karnataka India Department of Electrical and Electronics Engineering Prince Shri Venkateshwara Padmavathy Engineering College Chennai Department of Computer Science & Engineering Vivekananda Global University Jaipur India

herbal language processing (NLP) is an unexpectedly advancing field state-of-the-art synthetic intelligence and system ultra-modern. It's been able to permit computer systems to apprehend human language and to perform numerous tasks, from textual content type to automated summarization. Automatic reasoning is one of the most advanced and powerful packages modern-day NLP that promises to unlock the electricity cutting-edge computer systems to recognize and answer questions posed in herbal language. Automatic reasoning systems can analyze a fixed ultra-modern statistics after which reason logically and efficaciously based at the information supplied. This paper will speak the demanding situations latest enforcing automated reasoning, and the way improvements in NLP are getting used to enhance the accuracy and pace ultra-modern automatic reasoning. It'll additionally explore the packages contemporary automated reasoning in various domain names, along with question answering and choice making. finally, it'll discuss the capability for automatic reasoning to revolutionize the sector cutting-edge natural language processing by means of advancing the in a ramification modern methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Task groupings regularization: data-free meta-learning with heterogeneous pre-trained models 24

Task groupings regularization: data-free meta-learning with ...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Yongxian Wei Zixuan Hu Li Shen Zhenyi Wang Yu Li Chun Yuan Dacheng Tao Tsinghua Shenzhen International Graduate School Tsinghua University China Tsinghua Shenzhen International Graduate School Tsinghua University China and College of Computing & Data Science Nanyang Technological University Singapore School of Cyber Science and Technology Sun Yatsen University China University of Maryland College Park Department of Computer Science and Engineering The Chinese University of Hong Kong China College of Computing & Data Science Nanyang Technological University Singapore

data-Free Meta-Learning (DFML) aims to derive knowledge from a collection of pre-trained models without accessing their original data, enabling the rapid adaptation to new unseen tasks. Current methods often overlook the heterogeneity among pre-trained models, which leads to performance degradation due to task conflicts. In this paper, we empirically and theoretically identify and analyze the model heterogeneity in DFML. We find that model heterogeneity introduces a heterogeneity-homogeneity trade-off, where homogeneous models reduce task conflicts but also increase the overfitting risk. Balancing this trade-off is crucial for learning shared representations across tasks. Based on our findings, we propose Task Groupings Regularization, a novel approach that benefits from model heterogeneity by grouping and aligning conflicting tasks. Specifically, we embed pre-trained models into a task space to compute dissimilarity, and group heterogeneous models together based on this measure. Then, we introduce implicit gradient regularization within each group to mitigate potential conflicts. By encouraging a gradient direction suitable for all tasks, the meta-model captures shared representations that generalize across tasks. Comprehensive experiments showcase the superiority of our approach in multiple benchmarks, effectively tackling the model heterogeneity in challenging multidomain and multi-architecture scenarios.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multimodal Learning Approach for Multi-Topic Twitter Summarization 7

Multimodal Learning Approach for Multi-Topic Twitter Summari...

引用

7th International Conference on Cloud Computing and Big data Analytics, ICCCBDA 2022

作者： Cao, Yangyang Wu, Yanmin Qi, Jinli Chen, Zizhong Chongqing Key Lab. of Compl. Intelligence Chongqing University of Posts and Telecommunications Chongqing400065 China ChongQing College of Electronic Engineering Department of Artificial Intelligence and Big Data ChongQing401331 China University of California Riverside Department of Computer Science and Engineering RiversideCA United States

ISBN: (纸本)9781665487115

Twitter summary is designed to filter out the content summary from lots of noisy tweets, which can be used for Twitter search, public opinion analysis, hot topic discovery, etc. But the existing Twitter summary methods mostly focus on individual topics or hot events, and there are few studies on Twitter summary with multi-Topic posts. For the sake of resolve this problem, we propose a Twitter summary method based on social network information and manifold learning, which can extract tweets covering multiple topics, and better integrates the topic information and structural information of the tweets. First, the multi-modal manifold learning method is improved, and the social network information is integrated. Second, in order to test the relevance of the tweet summary and the topic, a test method of topic membership is designed. What's more, due to the lack of evaluation data sets, the author built a manual evaluation dataset. Our experiments on large-scale real tweets demonstrate the effectiveness of our framework. © 2022 IEEE.

关键词： Cloud computing Social networking (online) Blogs Manuals Big data Manifold learning Noise measurement

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：