检索结果-内蒙古大学图书馆

11th International Conference on Learning Representations, ICLR 2023

作者： Chen, Yixuan Shi, Yubin Dong, Mingzhi Yang, Xiaochen Li, Dongsheng Wang, Yujiang Dick, Robert P. Lv, Qin Zhao, Yingying Yang, Fan Gu, Ning Shang, Li China and Shanghai Key Laboratory of Data Science School of Computer Science Fudan University Shanghai China School of Mathematics Statistics The University of Glasgow Glasgow United Kingdom Microsoft Research Asia Shanghai China Department of Engineering Science University of Oxford Oxford United Kingdom Department of Electrical Engineering and Computer Science University of Michigan Michigan United States Department of Computer Science University of Colorado Boulder BoulderCO United States School of Microelectronics Fudan University Shanghai China

This work pursues the optimization of over-parameterized deep models for superior training efficiency and test performance. We first theoretically emphasize the importance of two properties of over-parameterized models, i.e., the convergence gap and the generalization gap. Subsequent analyses unveil that these two gaps can be upper-bounded by the ratio of the Lipschitz constant and the Polyak-Lojasiewicz (PL) constant, a crucial term abbreviated as the condition number. Such discoveries have led to a structured pruning method with a novel pruning criterion. That is, we devise a gating network that dynamically detects and masks out those poorly-behaved nodes of a deep model during the training session. To this end, this gating network is learned via minimizing the condition number of the target model, and this process can be implemented as an extra regularization loss term. Experimental studies demonstrate that the proposed method outperforms the baselines in terms of both training efficiency and test performance, exhibiting the potential of generalizing to a variety of deep network architectures and tasks. © 2023 11th International Conference on Learning Representations, ICLR 2023. All rights reserved.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Multiband Kick Drum Separation with Genre-Guided Optimization 10th

Multiband Kick Drum Separation with Genre-Guided Optimizati...

引用

10th Conference on Sound and Music Technology, CSMT 2023

作者： Li, Si Jiang, Yuanfeng Wang, Zhaowen Qian, Jiale Li, Wei School of Computer Science and Technology Fudan University Shanghai200438 China Department of Music AI and Information Technology Central Conservatory of Music Beijing100091 China Shanghai Key Laboratory of Intelligent Information Processing Fudan University Shanghai200433 China

ISBN: (纸本)9789819779611

Kick drum plays an important role in electronic music. When hearing an excellent kick drum sound from a reference music track, music production beginners always feel confused about how it will sound if we solo the kick track and why it sounds perfect in a specific song. Reproducing the kick of a music track is not easy even for professional producers and arrangers. To investigate the ability of artificial intelligence in separating kick-drum sounds, in this paper we propose an end-to-end system which is capable of separating multi-genre high-quality electronic kick drum clips from mixed song tracks. Unlike most of the previous music source separation tasks, we only focus on the electronic kick drum rather than the whole percussion track. First, we design a multiband data representation approach for non-pitched percussions such as kick drums, then adopt a multi-discriminator module trained on a multi-genre kick drum corpus to further optimize the separated kick sound. To the authors’ knowledge, this study marks the first attempt at kick drum separation with stylistic optimization. Objective and subjective evaluations demonstrate that joint multiband representation and genre-guided optimisation can form a high-performance framework, showing a promising solution to the kick drum separation task. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Source separation

来源：评论

学校读者我要写书评

暂无评论

Service-Based Event Penetration from IoT Sensors to Businesses: a Case Study

Service-Based Event Penetration from IoT Sensors to Business...

引用

2022 CCF International Conference on Service science, CCF ICSS 2022

作者： Wang, Guiling Fang, Jun Wang, Jing Yu, Jian Zhang, Liang Han, Yanbo Beijing Key Laboratory on Integration and Analysis of Large-Scale Stream Data Beijing China North China University of Technology School of Information Science and Technology Beijing China Auckland University of Technology Department of Computer Science Auckland New Zealand Shanghai Key Laboratory of Data Science Shanghai China Shanghai Institute of Intelligent Electronics & Systems Shanghai China

ISBN: (数字)9781665498616

ISBN: (纸本)9781665498616

By leveraging IoT Big data, BPM can gain real-time physical world information to make faster and more accurate decisions, but there is a technical gap between IoT sensors and businesses. To bridge the gap, an event penetration mechanism from IoT sensors to business processes is proposed along a practical case study. This paper presents a concrete IoT-BPM application case dealing with seaborne safety ensurance in transporting liquefied natural gas (LNG), analyzes its technical challenges, and examines the feasibility and supposed effects of the BRIBOT approach. © 2022 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

MFCM-DTI model of multimodal feature fusion: Prediction of drug-target interaction

MFCM-DTI model of multimodal feature fusion: Prediction of d...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Li, Wenjun Ma, Wanjun Yang, Mengyun Tang, Xiwei Changsha University of Science and Technology Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation Changsha China Hunan First Normal University School of Computer Science Changsha China Hunan First Normal University School of Intelligent Manufacturing Changsha China

ISBN: (纸本)9798350386226

Drug repositioning is a vital area of biomedicine, where confirming interactions between drugs and specific targets is essential for establishing the efficacy of pharmaceutical agents. Traditional in vitro screening methods have limitations, prompting the use of computer simulations as an effective alternative for predicting drug-target interactions (DTI). This approach has gained significant attention in the scientific community. In this study, we introduce MFCM-DTI, a DTI prediction model that employs multimodal features to accurately capture the intricate interactions between drug molecular structures and key amino acids of target proteins. Our results demonstrate that MFCM-DTI outperforms existing models in prediction accuracy and robustness. Furthermore, MFCM-DTI has been successfully used to predict interactions between key SARS-CoV-2 proteins and existing drugs, providing a solid foundation for developing therapeutic agents against SARS-CoV-2 infection. This study underscores the broad applicability and strong predictive capabilities of MFCM-DTI in drug-protein interaction prediction, opening new avenues for research. The predicted drug targets and interaction data offer valuable insights for future experimental validation and clinical trials, potentially driving innovation in the biomedical field. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

FedLMT: Tackling System Heterogeneity of Federated Learning via Low-Rank Model Training with Theoretical Guarantees 41

FedLMT: Tackling System Heterogeneity of Federated Learning ...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Liu, Jiahao Zhou, Yipeng Wu, Di Hu, Miao Guizani, Mohsen Sheng, Quan Z. School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China School of Computing Faculty of Science and Engineering Macquarie University Sydney Australia Abu Dhabi United Arab Emirates

Federated learning (FL) is an emerging machine learning paradigm for preserving data privacy. However, diverse client hardware often has varying computation resources. Such system heterogeneity limits the participation of resource-constrained clients in FL, and hence degrades the global model accuracy. To enable heterogeneous clients to participate in and contribute to FL training, previous works tackle this problem by assigning customized sub-models to individual clients with model pruning, distillation, or low-rank based techniques. Unfortunately, the global model trained by these methods still encounters performance degradation due to heterogeneous sub-model aggregation. Besides, most methods are heuristic-based and lack convergence analysis. In this work, we propose the FedLMT framework to bridge the performance gap, by assigning clients with a homogeneous pre-factorized low-rank model to substantially reduce resource consumption without conducting heterogeneous aggregation. We theoretically prove that the convergence of the low-rank model can guarantee the convergence of the original full model. To further meet clients' personalized resource needs, we extend FedLMT to pFedLMT, by separating model parameters into common and custom ones. Finally, extensive experiments are conducted to verify our theoretical analysis and show that FedLMT and pFedLMT outperform other baselines with much less communication and computation costs. Copyright 2024 by the author(s)

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

CLIDSUM: A Benchmark dataset for Cross-Lingual Dialogue Summarization

CLIDSUM: A Benchmark Dataset for Cross-Lingual Dialogue Summ...

引用

2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022

作者： Wang, Jiaan Meng, Fandong Lu, Ziyao Zheng, Duo Li, Zhixu Qu, Jianfeng Zhou, Jie Pattern Recognition Center WeChat AI Tencent Inc China School of Computer Science and Technology Soochow University Suzhou China Shanghai Key Laboratory of Data Science School of Computer Science Fudan University Shanghai China Beijing University of Posts and Telecommunications Beijing China

We present CLIDSUM, a benchmark dataset towards building cross-lingual summarization systems on dialogue documents. It consists of 67k+ dialogue documents and 112k+ annotated summaries in different target languages. Based on the proposed CLIDSUM, we introduce two benchmark settings for supervised and semi-supervised scenarios, respectively. We then build various baseline systems in different paradigms (pipeline and end-to-end) and conduct extensive experiments on CLIDSUM to provide deeper analyses. Furthermore, we propose mDIALBART which extends mBART via further pre-training, where the multiple objectives help the pre-trained model capture the structural characteristics as well as key content in dialogues and the transformation from source to the target language. Experimental results show the superiority of mDIALBART, as an end-to-end model, outperforms strong pipeline models on CLIDSUM. Finally, we discuss specific challenges that current approaches faced with this task and give multiple promising directions for future research. We have released the dataset and code at https://***/krystalan/ClidSum. © 2022 Association for Computational Linguistics.

关键词： Pipelines

来源：评论

学校读者我要写书评

暂无评论

Trajectory set Empowered Hypergraph Transformer for Mobile Sensor Based Traffic Prediction

Trajectory set Empowered Hypergraph Transformer for Mobile S...

引用

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

作者： Hanyuan Zhang Xinyu Zhang Qize Jiang Liang Li Baihua Zheng Weiwei Sun School of Computer Science Shanghai Key Laboratory of Data Science Fudan University Shanghai Institute of Intelligent Electronics & Systems School of Computing and Information Systems Singapore Management University

Traffic speed prediction is vital for intelligent transportation systems. However, most existing methods focus on costly static sensors. In contrast, utilizing GPS devices from vehicles as mobile sensors offers a cost-effective means to gather dynamic traffic data. Despite the presence of historical trajectory data, mobile sensor-based traffic prediction remains under-explored. Existing methods often treat trajectories as substitutes for static sensors, missing the full utilization of the spatial-temporal signals within the complete trajectory set. To address this, we propose TrajHGT, a novel trajectory set empowered hypergraph transformer model that captures trafficrelated spatial-temporal features through adaptive attention and fusion mechanisms in both the trajectory hypergraph space and the road graph space. Real dataset experiments demonstrate the superiority of TrajHGT.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A versatile framework for analyzing galaxy image data by incorporating Human-in-the-loop in a large vision model

引用

Chinese Physics C 2024年第9期48卷 176-187页

作者： Ming-Xiang Fu Yu Song Jia-Meng Lv Liang Cao Peng Jia Nan Li Xiang-Ru Li Ji-Feng Liu A-Li Luo Bo Qiu Shi-Yin Shen Liang-Ping Tu Li-Li Wang Shou-Lin Wei Hai-Feng Yang Zhen-Ping Yi Zhi-Qiang Zou National Astronomical Observatories Chinese Academy of SciencesBeijing 100101China College of Electronic Information and Optical Engineering Taiyuan University of TechnologyTaiyuan 030024China School of Astronomy and Space Science University of Chinese Academy of SciencesBeijing 101408China Key lab of Space Astronomy and Technology National Astronomical ObservatoriesBeijing 100101China School of Computer Science South China Normal UniversityGuangzhou 510631China CAS Key Laboratory of Optical Astronomy National Astronomical ObservatoriesBeijing 100101China University of Chinese Academy of Sciences NanjingNanjing 211135China University of Science and Technology Beijing Beijing 100083China Shanghai Astronomical Observatory Chinese Academy of SciencesShanghai 200030China School of Computer and Information Dezhou UniversityDezhou 253023China Faculty of Information Engineering and Automation Kunming University of Science and TechnologyKunming 650500China School of Computer Science and Technology Taiyuan University of Science and TechnologyTaiyuan 30024China School of Mechanical Electrical and Information EngineeringShandong UniversityWeihai 264209China Nanjing University of Posts&Telecommunications Nanjing 210023China School of Science University of Science and Technology LiaoningAnshan 114051China

The exponential growth of astronomical datasets provides an unprecedented opportunity for humans to gain insight into the ***,effectively analyzing this vast amount of data poses a significant *** response,astronomers are turning to deep learning techniques,but these methods are limited by their specific training sets,leading to considerable duplicate *** overcome this issue,we built a framework for the general analysis of galaxy images based on a large vision model(LVM)plus downstream tasks(DST),including galaxy morphological classification,image restoration object detection,parameter extraction,and *** the low signal-to-noise ratios of galaxy images and the imbalanced distribution of galaxy categories,we designed our LVM to incorporate a Human-in-the-loop(HITL)module,which leverages human knowledge to enhance the reliability and interpretability of processing galaxy images *** proposed framework exhibits notable fewshot learning capabilities and versatile adaptability for all the abovementioned tasks on galaxy images in the DESI Legacy Imaging *** particular,for the object detection task,which was trained using 1000 data points,our DST in the LVM achieved an accuracy of 96.7%,while ResNet50 plus Mask R-CNN reached an accuracy of 93.1%.For morphological classification,to obtain an area under the curve(AUC)of~0.9,LVM plus DST and HITL only requested 1/50 of the training sets that ResNet18 *** addition,multimodal data can be integrated,which creates possibilities for conducting joint analyses with datasets spanning diverse domains in the era of multi-messenger astronomy.

关键词： artificial intelligence large vision model human-in-the-loop astronomy galaxies

来源：评论

学校读者我要写书评

暂无评论

SoVAR: Build Generalizable Scenarios from Accident Reports for Autonomous Driving Testing 24

SoVAR: Build Generalizable Scenarios from Accident Reports f...

引用

39th ACM/IEEE International Conference on Automated Software Engineering, ASE 2024

作者： Guo, An Zhou, Yuan Tian, Haoxiang Fang, Chunrong Sun, Yunjian Sun, Weisong Gao, Xinyu Luu, Anh Tuan Liu, Yang Chen, Zhenyu State Key Laboratory for Novel Software Technology Nanjing University Nanjing210023 China School of Computer Science and Technology Zhejiang Sci-Tech University Hangzhou310018 China College of Computing and Data Science Nanyang Technological University Singapore

ISBN: (纸本)9798400712487

Autonomous driving systems (ADSs) have undergone remarkable development and are increasingly employed in safety-critical applications. However, recently reported data on fatal accidents involving ADSs suggests that the desired level of safety has not yet been fully achieved. Consequently, there is a growing need for more comprehensive and targeted testing approaches to ensure safe driving. Scenarios from real-world accident reports provide valuable resources for ADS testing, including critical scenarios and high-quality seeds. However, existing scenario reconstruction methods from accident reports often exhibit limited accuracy in information extraction. Moreover, due to the diversity and complexity of road environments, matching current accident information with the simulation map data for reconstruction poses significant *** this paper, we design and implement SoVAR, a tool for automatically generating road-generalizable scenarios from accident reports. SoVAR utilizes well-designed prompts with linguistic patterns to guide the large language model (LLM) in extracting accident information from textual data. Subsequently, it formulates and solves accident-related constraints in conjunction with the extracted accident information to generate accident trajectories. Finally, SoVAR reconstructs accident scenarios on various map structures and converts them into test scenarios to evaluate its capability to detect defects in industrial ADSs. We experiment with SoVAR, using the accident reports from the National Highway Traffic Safety Administration's (NHTSA) database to generate test scenarios for the industrial-grade ADS Apollo. The experimental findings demonstrate that SoVAR can effectively generate generalized accident scenarios across different road structures. Furthermore, the results confirm that SoVAR identified 5 distinct safety violation types that contributed to the crash of Baidu Apollo. © 2024 Copyright is held by the owner/author(s). Publication ri

关键词： Highway administration

来源：评论

学校读者我要写书评

暂无评论

HAUSER: Towards Holistic and Automatic Evaluation of Simile Generation

arXiv

引用

arXiv 2023年

作者： He, Qianyu Zhang, Yikai Liang, Jiaqing Huang, Yuncheng Xiao, Yanghua Chen, Yunwen Shanghai Key Laboratory of Data Science School of Computer Science Fudan University China School of Data Science Fudan University China DataGrand Inc. Shanghai China

Similes play an imperative role in creative writing such as story and dialogue generation. Proper evaluation metrics are like a beacon guiding the research of simile generation (SG). However, it remains under-explored as to what criteria should be considered, how to quantify each criterion into metrics, and whether the metrics are effective for comprehensive, efficient, and reliable SG evaluation. To address the issues, we establish HAUSER, a holistic and automatic evaluation system for the SG task, which consists of five criteria from three perspectives and automatic metrics for each criterion. Through extensive experiments, we verify that our metrics are significantly more correlated with human ratings from each perspective compared with prior automatic metrics. Resources of HAUSER are publicly available at https://***/Abbey4799/HAUSER. Copyright © 2023, The Authors. All rights reserved.

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：