检索结果-内蒙古大学图书馆

Recommender systems based on generative adversarial networks: a problem-driven perspective

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Gao, Min Zhang, Junwei Yu, Junliang Li, Jundong Wen, Junhao Xiong, Qingyu Department of Electrical and Computer Engineering Department of Computer Science School of Data Science University of Virginia Charlottesville22904 United States School of Information Technology and Electrical Engineering University of Queensland QLD4072 Australia School of Big Data and Software Engineering Chongqing University Chongqing401331 China

Recommender systems (RS) play a very important role in various aspects of people’s online life. Many companies leverage RS to help users discover new and favored items. Despite their empirical success, these systems still suffer from two main problems: data noise and data sparsity. In recent years, Generative Adversarial Networks (GANs) have received a surge of interests in many fields because of their great potential to learn complex real data distribution, and they also provide new means to mitigate the aforementioned problems of RS. Particularly, owing to adversarial learning, the problem of data noise can be handled by adding adversarial perturbations or forcing discriminators to tell the informative and uninformative data examples apart. As for the mitigation of data sparsity issue, the GAN-based models are able to replicate the real distribution of the user-item interactions and augment the available data. To gain a comprehensive understanding of these GAN-based recommendation models, we provide a retrospective of these studies and organize them from a problem-driven perspective. Specifically, we propose a taxonomy of these models, along with a detailed description of them and their advantages. Finally, we elaborate on several open issues and expand on current trends in the GAN-based RS. Copyright © 2020, The Authors. All rights reserved.

关键词： Generative adversarial networks

Semantics Analytics of Origin-Destination Flows from Crowd Sensed Big data

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2019年第7期61卷 227-241页

作者： Ning Cao Shengfang Li Keyong Shen Sheng Bin Gengxin Sun Dongjie Zhu Xiuli Han Guangsheng Cao Abraham Campbell College of Computer Information and Engineering Nanchang Institute of TechnologyNanchangChina College of Information Engineering Sanming UniversitySanmingChina School of Data Science and Software Engineering Qingdao UniversityQingdaoChina School of Computer Science and Technology Harbin Institute of TechnologyWeihaiChina Public Teaching Department Qingdao Technical CollegeQingdaoChina School of Computer Science University College DublinDublinIreland

Monitoring,understanding and predicting Origin-destination(OD)flows in a city is an important problem for city planning and human ***-GPS traces,acted as one kind of typical crowd sensed data,it can be used to mine the semantics of OD *** this paper,we firstly construct and analyze a complex network of OD flows based on large-scale GPS taxi traces of a city in *** spatiotemporal analysis for the OD flows complex network showed that there were distinctive patterns in OD *** based on a novel complex network model,a semantics mining method of OD flows is proposed through compounding Points of Interests(POI)network and public transport network to the OD flows *** propose method would offer a novel way to predict the location characteristic and future traffic conditions accurately.

关键词： Origin-destination(OD)flows semantics analytics complex network big data analysis

Combinatorial Optimization and Reasoning with Graph Neural Networks

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Cappart, Quentin Chételat, Didier Khalil, Elias B. Lodi, Andrea Morris, Christopher Veličković, Petar Department of Computer Engineering and Software Engineering Polytechnique Montréal Montréal Canada CERC in Data Science for Real-Time Decision-Making Polytechnique Montréal Montréal Canada Department of Mechanical & Industrial Engineering University of Toronto Toronto Canada Jacobs Technion-Cornell Institute Cornell Tech and Technion - IIT New York United States Department of Computer Science RWTH Aachen University Aachen Germany DeepMind London United Kingdom

Combinatorial optimization is a well-established area in operations research and computer science. Until recently, its methods have focused on solving problem instances in isolation, ignoring that they often stem from related data distributions in practice. However, recent years have seen a surge of interest in using machine learning, especially graph neural networks (GNNs), as a key building block for combinatorial tasks, either directly as solvers or by enhancing exact solvers. The inductive bias of GNNs effectively encodes combinatorial and relational input due to their invariance to permutations and awareness of input sparsity. This paper presents a conceptual review of recent key advancements in this emerging field, aiming at optimization and machine learning researchers. Copyright © 2021, The Authors. All rights reserved.

关键词： Combinatorial optimization

Mask R-CNN with R-D Net: A Detection and Segmentation Model for Early Gastric Cancer Gastroscopy Images

学校读者我要写书评

暂无评论

Mask R-CNN with R-D Net: A Detection and Segmentation Model ...

Electronic Information engineering and Computer science (EIECS), 2021 International Conference on

作者： Tao Liang Yi Lei Wenai Song Xue Sun Jun Du Qing Wang Li Zhao School of Software Engineering North University of China Taiyuan China Faculty of Information Technology Beijing University of Technology Beijing China VIP Department and Family Medicine Department Beijing Hospital National Center of Gerontology Institute of Geriatric Medicine Chinese Academy of Medical Sciences Beijing China Department of Pathology Beijing Hospital National Center of Gerontology Institute of Geriatric Medicine Chinese Academy of Medical Sciences Beijing China Pharmacovigilance Research Center for Information Technology and Data Science Cross-Strait Tsinghua Research Institute Xiamen China Department of Automation Tsinghua University Beijing China Department of Gastroenterology Beijing Hospital National Center of Gerontology Institute of Geriatric Medicine Chinese Academy of Medical Sciences Beijing China

Gastric cancer is one of the most common malignant tumors of the digestive system. Although gastroscopy can effectively detect early cancers, manual identification and diagnosis is time-consuming and laborious, and the false-negative rate is as high as 4.6-25.8%. Early gastric cancer lesion areas can be detected and segmented using deep learning techniques. We collected and organized 12665 eligible gastroscopy images, and cropped and expanded the original dataset after labeling by professional doctors, and finally obtained 19673 images for model training. We propose ResMSE Net and DWFF Net and apply these two networks to Mask R-CNN to obtain a new model: Mask R-CNN With R-D Net. On the same standard dataset, the mAP@[IoU=0.5] and ACC values of the Mask R-CNN With R-D Net model achieve respectively 45.37% and 67.74%, which are improved by 5.83% and 9.08%, respectively, compared with the original model. In our experiments, we found that Mask R-CNN With R-D Net is difficult to detect targets that are extremely similar to the background, which is the direction of our future work. Secondly, we hope to minimize the number of model parameters and make the model more lightweight under the premise of ensuring accuracy, so that the model can be embedded into gastroscopy detection equipment for real-time detection.

关键词：

Improving Transformers with Probabilistic Attention Keys

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Nguyen, Tam Nguyen, Tan M. Le, Dung D. Nguyen, Duy Khuong Tran, Viet-Anh Baraniuk, Richard G. Ho, Nhat Osher, Stanley J. FPT Software AI Center Hanoi Viet Nam Department of Mathematics University of California Los AngelesCA United States College of Engineering and Computer Science VinUniversity Hanoi Viet Nam Deezer Research France Department of Electrical and Computer Engineering Rice University HoustonTX United States Department of Statistics and Data Sciences The University of Texas AustinTX United States

Multi-head attention is a driving force behind state-of-the-art transformers, which achieve remarkable performance across a variety of natural language processing (NLP) and computer vision tasks. It has been observed that for many applications, those attention heads learn redundant embedding, and most of them can be removed without degrading the performance of the model. Inspired by this observation, we propose Transformer with a Mixture of Gaussian Keys (Transformer-MGK), a novel transformer architecture that replaces redundant heads in transformers with a mixture of keys at each head. These mixtures of keys follow a Gaussian mixture model and allow each attention head to focus on different parts of the input sequence efficiently. Compared to its conventional transformer counterpart, Transformer-MGK accelerates training and inference, has fewer parameters, and requires fewer FLOPs to compute while achieving comparable or better accuracy across tasks. Transformer-MGK can also be easily extended to use with linear attention. We empirically demonstrate the advantage of Transformer-MGK in a range of practical applications, including language modeling and tasks that involve very long sequences. On the Wikitext-103 and Long Range Arena benchmark, Transformer-MGKs with 4 heads attain comparable or better performance to the baseline transformers with 8 heads. © 2021, CC BY.

关键词： Benchmarking

An Audio-Visual dataset and Deep Learning Frameworks for Crowded Scene Classification

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Pham, Lam Ngo, Dat Nguyen, Phu X. Hoang, Truong Schindler, Alexander Competence Unit Data Science & Artificial Intelligence Center for Digital Safety & Security Austria Institute of Technology Austria School of Computer Science and Electronic Engineering University of Essex United Kingdom Department of Computer Fundamentals FPT University Ho Chi Minh City700000 Viet Nam FPT Software Company Limited Viet Nam

This paper presents a task of audio-visual scene classification (SC) where input videos are classified into one of five real-life crowded scenes: 'Riot', 'Noise-Street', 'Firework-Event', 'Music-Event', and 'Sport-Atmosphere'. To this end, we firstly collect an audio-visual dataset (videos) of these five crowded contexts from Youtube (in-the-wild scenes). Then, a wide range of deep learning frameworks are proposed to deploy either audio or visual input data independently. Finally, results obtained from high-performed deep learning frameworks are fused to achieve the best accuracy score. Our experimental results indicate that audio and visual input factors independently contribute to the SC task's performance. Significantly, an ensemble of deep learning frameworks exploring either audio or visual input data can achieve the best accuracy of 95.7%. Copyright © 2021, The Authors. All rights reserved.

关键词： Deep learning

KFR: Optimal Cache Management with K-Framed Reclamation for Drive-Managed SMR Disks

学校读者我要写书评

暂无评论

KFR: Optimal Cache Management with K-Framed Reclamation for ...

Design Automation Conference

作者： Chenlin Ma Yi Wang Zhaoyan Shen Zili Shao College of Big Data and Internet Shenzhen Technology University College of Computer Science and Software Engineering Shenzhen University School of Computer Science and Technology Shandong University Department of Computer Science and Engineering The Chinese University of Hong Kong

ISBN: (数字)9781728110851

ISBN: (纸本)9781728158020

Shingled Magnetic Recording (SMR) disks have been proposed as a promising solution to satisfy the increasing capacity need in the big data era. Drive-Managed SMR (DM-SMR) disk which acts as a traditional block device is favored for providing high compatibility. However, DM-SMR disks suffer from high performance recovery time (PRT) due to the "SMR space reclamation" issue. This paper proposes an optimal cache management named K-Framed Reclamation (KFR) to minimize PRT within the DM-SMR disk. The effectiveness of our proposed design was evaluated with realistic and intensive I/O workloads and the results are encouraging.

关键词： Delays Magnetic recording Writing Big data Computer science Performance evaluation Runtime

A Consensus Privacy Metrics Framework for Synthetic data

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Pilgram, Lisa Dankar, Fida K. Drechsler, Jörg Elliot, Mark Domingo-Ferrer, Josep Francis, Paul Kantarcioglu, Murat Kong, Linglong Malin, Bradley Muralidhar, Krishnamurty Myles, Puja Prasser, Fabian Raisaro, Jean Louis Yan, Chao El Emam, Khaled School of Epidemiology and Public Health University of Ottawa ON Canada CHEO Research Institute ON Canada Department of Nephrology and Medical Intensive Care Charité – Universitätsmedizin Berlin Berlin Germany Department for Statistical Methods Institute for Employment Research Nuernberg Germany Institute for Statistics Ludwig-Maximilians-Universität Munich Germany Joint Program in Survey Methodology University of Maryland United States The Cathie Marsh Institute Research School of Social Sciences University of Manchester Manchester United Kingdom Department of Computer Engineering and Mathematics Universitat Rovira i Virgili Catalonia Tarragona Spain Max Planck Institute for Software Systems Germany Department of Computer Science Virginia Tech United States Department of Mathematical and Statistical Sciences University of Alberta Alberta Canada Department of Biomedical Informatics Vanderbilt University Medical Center NashvilleTN United States Department of Biostatistics Vanderbilt University Medical Center NashvilleTN United States Department of Computer Science Vanderbilt University NashvilleTN United States Department of Marketing and Supply Chain Management University of Oklahoma Oklahoma United States Medicines and Healthcare products Regulatory Agency London United Kingdom Berlin Institute of Health at Charité Universitätsmedizin Berlin Medical Informatics Group Berlin Germany Biomedical Data Science Center University Hospital Lausanne Lausanne Switzerland

Synthetic data generation is one approach for sharing individual-level data. However, to meet legislative requirements, it is necessary to demonstrate that the individuals’ privacy is adequately protected. There is no consolidated standard for measuring privacy in synthetic data. Through an expert panel and consensus process, we developed a framework for evaluating privacy in synthetic data. Our findings indicate that current similarity metrics fail to measure identity disclosure, and their use is discouraged. For differentially private synthetic data, a privacy budget other than close to zero was not considered interpretable. There was consensus on the importance of membership and attribute disclosure, both of which involve inferring personal information about an individual without necessarily revealing their identity. The resultant framework provides precise recommendations for metrics that address these types of disclosures effectively. Our findings further present specific opportunities for future research that can help with widespread adoption of synthetic data. © 2025, CC BY.

关键词： Differential privacy

Calling Cards: A Customizable Platform to Longitudinally Record Protein-DNA Interactions Over Time in Cells and Tissues (vol 3, e883, 2023)

学校读者我要写书评

暂无评论

CURRENT PROTOCOLS

CURRENT PROTOCOLS 2023年第10期3卷 e883页

作者： Yen, Allen Mateusiak, Chase Sarafinovska, Simona Gachechiladze, Mariam A. Guo, Juanru Chen, Xuhua Moudgil, Arnav Cammack, Alexander J. Hoisington-Lopez, Jessica Crosby, MariaLynn Brent, Michael R. Mitra, Robi D. Dougherty, Joseph D. Department of Genetics Washington University School of Medicine Saint Louis Missouri Department of Psychiatry Washington University School of Medicine Saint Louis Missouri Contribution: Conceptualization Data curation Formal analysis Funding acquisition Investigation Methodology Project administration Software Supervision Validation Visualization Writing - original draft Writing - review & editing Edison Family Center for Genome Sciences and Systems Biology Washington University School of Medicine Saint Louis Missouri Contribution: Conceptualization Data curation Formal analysis Methodology Software Visualization Writing - original draft Contribution: Formal analysis Funding acquisition Investigation Visualization Writing - original draft Writing - review & editing Contribution: Formal analysis Funding acquisition Investigation Visualization Writing - review & editing Contribution: Data curation Methodology Software Writing - review & editing Contribution: Methodology Writing - review & editing Current address: Department of Ophthalmology Stanford University School of Medicine Stanford California Contribution: Conceptualization Methodology Software Validation Department of Neurology Washington University School of Medicine Saint Louis Missouri Current address: Department of Neurodegenerative Diseases University College London London United Kingdom Contribution: Conceptualization Methodology Validation Contribution: Investigation Writing - review & editing Department of Computer Science and Engineering Washington University Saint Louis Missouri Contribution: Funding acquisition Contribution: Conceptualization Funding acquisition Project administration Resources Software Supervision Writing - review & editing Intellectual and Developmental Disabilities Research Center Washington University School of Medicine Saint Louis Missouri Contribution: Conceptualization Funding acquisition Project administration Resources Supervision Writing - origin

Calling Cards is a platform technology to record a cumulative history of transient protein-DNA interactions in the genome of genetically targeted cell types. The record of these interactions is recovered by next-generation sequencing. Compared with other genomic assays, readouts of which provide a snapshot at the time of harvest, Calling Cards enables correlation of historical molecular states to eventual outcomes or phenotypes. To achieve this, Calling Cards uses the piggyBac transposase to insert self-reporting transposon “Calling Cards” into the genome, leaving permanent marks at interaction sites. Calling Cards can be deployed in a variety of in vitro and in vivo biological systems to study gene regulatory networks involved in development, aging, and disease. Out of the box, it assesses enhancer usage but can be adapted to profile-specific transcription factor (TF) binding with custom TF-piggyBac fusion proteins. The Calling Cards workflow has five main stages: delivery of Calling Cards reagents, sample preparation, library preparation, sequencing, and data analysis. Here, we first present a comprehensive guide for experimental design, reagent selection, and optional customization of the platform to study additional TFs. Then, we provide an updated protocol for the five steps, using reagents that improve throughput and decrease costs, including an overview of a newly deployed computational pipeline. This protocol is designed for users with basic molecular biology experience to process samples into sequencing libraries in 2 days. Familiarity with bioinformatic analysis and command line tools is required to set up the pipeline in a high-performance computing environment and to conduct downstream analyses. © 2023 Wiley Periodicals LLC. This article was corrected on 16 October 2023. See the end of the full text for details. Basic Protocol 1 : Preparation and delivery of Calling Cards reagents Support Protocol 1 : Next-generation sequencing quantification of barcode

关键词： calling cards epigenetics enhancer usage gene regulation transcription factor binding