检索结果-内蒙古大学图书馆

Not All Diffusion Model Activations Have Been Evaluated as Discriminative Features

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Meng, Benyuan Xu, Qianqian Wang, Zitai Cao, Xiaochun Huang, Qingming Institute of Information Engineering CAS China School of Cyber Security University of Chinese Academy of Sciences China Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS China Peng Cheng Laboratory China School of Cyber Science and Tech. Shenzhen Campus of Sun Yat-sen University China School of Computer Science and Tech. University of Chinese Academy of Sciences China Key Laboratory of Big Data Mining and Knowledge Management CAS China

Diffusion models are initially designed for image generation. Recent research shows that the internal signals within their backbones, named activations, can also serve as dense features for various discriminative tasks such as semantic segmentation. Given numerous activations, selecting a small yet effective subset poses a fundamental problem. To this end, the early study of this field performs a large-scale quantitative comparison of the discriminative ability of the activations. However, we find that many potential activations have not been evaluated, such as the queries and keys used to compute attention scores. Moreover, recent advancements in diffusion architectures bring many new activations, such as those within embedded ViT modules. Both combined, activation selection remains unresolved but overlooked. To tackle this issue, this paper takes a further step with a much broader range of activations evaluated. Considering the significant increase in activations, a full-scale quantitative comparison is no longer operational. Instead, we seek to understand the properties of these activations, such that the activations that are clearly inferior can be filtered out in advance via simple qualitative evaluation. After careful analysis, we discover three properties universal among diffusion models, enabling this study to go beyond specific models. On top of this, we present effective feature selection solutions for several popular diffusion models. Finally, the experiments across multiple discriminative tasks validate the superiority of our method over the SOTA competitors. Our code is available at this url. © 2024, CC BY.

关键词： Semantics

Suppress Content Shift: Better Diffusion Features via Off-the-Shelf Generation Techniques

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Meng, Benyuan Xu, Qianqian Wang, Zitai Yang, Zhiyong Cao, Xiaochun Huang, Qingming Institute of Information Engineering CAS China School of Cyber Security University of Chinese Academy of Sciences China Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS China Peng Cheng Laboratory China School of Computer Science and Tech. University of Chinese Academy of Sciences China Key Laboratory of Big Data Mining and Knowledge Management CAS China School of Cyber Science and Tech. Sun Yat-sen University Shenzhen Campus China

Diffusion models are powerful generative models, and this capability can also be applied to discrimination. The inner activations of a pre-trained diffusion model can serve as features for discriminative tasks, namely, diffusion feature. We discover that diffusion feature has been hindered by a hidden yet universal phenomenon that we call content shift. To be specific, there are content differences between features and the input image, such as the exact shape of a certain object. We locate the cause of content shift as one inherent characteristic of diffusion models, which suggests the broad existence of this phenomenon in diffusion feature. Further empirical study also indicates that its negative impact is not negligible even when content shift is not visually perceivable. Hence, we propose to suppress content shift to enhance the overall quality of diffusion features. Specifically, content shift is related to the information drift during the process of recovering an image from the noisy input, pointing out the possibility of turning off-the-shelf generation techniques into tools for content shift suppression. We further propose a practical guideline named GATE to efficiently evaluate the potential benefit of a technique and provide an implementation of our methodology. Despite the simplicity, the proposed approach has achieved superior results on various tasks and datasets, validating its potential as a generic booster for diffusion features. Our code is available at this url. © 2024, CC BY.

关键词：

Joint Compression and Deadline Optimization for Wireless Federated Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhang, Maojun Li, Yang Liu, Dongzhu Jin, Richeng Zhu, Guangxu Zhong, Caijun Quek, Tony Q.S. The College of information Science and Electronic Engineering Zhejiang University Hangzhou China China Academy of Information and Communications Technology Beijing China School of Computing Science University of Glasgow United Kingdom The Department of Information and Communication Engineering Zhejiang University Hangzhou310007 China The Zhejiang–Singapore Innovation and AI Joint Research Lab Hangzhou310007 China Hangzhou310007 China The Shenzhen Research Institute of Big Data Guangdong Shenzhen518172 China The Peng Cheng Laboratory Guangdong Shenzhen518055 China Guangdong Guangzhou510555 China The Singapore University of Technology and Design Singapore487372 Singapore

Federated edge learning (FEEL) is a popular distributed learning framework for privacy-preserving at the edge, in which densely distributed edge devices periodically exchange model-updates with the server to complete the global model training. Due to limited bandwidth and uncertain wireless environment, FEEL may impose heavy burden to the current communication system. In addition, under the common FEEL framework, the server needs to wait for the slowest device to complete the update uploading before starting the aggregation process, leading to the straggler issue that causes prolonged communication time. In this paper, we propose to accelerate FEEL from two aspects: i.e., 1) performing data compression on the edge devices and 2) setting a deadline on the edge server to exclude the straggler devices. However, undesired gradient compression errors and transmission outage are introduced by the aforementioned operations respectively, affecting the convergence of FEEL as well. In view of these practical issues, we formulate a training time minimization problem, with the compression ratio and deadline to be optimized. To this end, an asymptotically unbiased aggregation scheme is first proposed to ensure zero optimality gap after convergence, and the impact of compression error and transmission outage on the overall training time are quantified through convergence analysis. Then, the formulated problem is solved in an alternating manner, based on which, the novel joint compression and deadline optimization (JCDO) algorithm is derived. Numerical experiments for different use cases in FEEL including image classification and autonomous driving show that the proposed method is nearly 30X faster than the vanilla FedSGD algorithm, and outperforms the state-of-the-art schemes. © 2023, CC BY.

关键词： Autonomous vehicles

Cheetah: Accelerating Dynamic Graph Mining with Grouping Updates

学校读者我要写书评

暂无评论

ACM Transactions on Architecture and Code Optimization 1000年

作者： Yi Zhang Xiaomeng Yi Yu Huang Jingrui Yuan Chuangyi Gui Dan Chen Long Zheng Jianhui Yue Xiaofei Liao Hai Jin Jingling Xue National Engineering Research Center for Big Data Technology and System Service Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China Zhejiang Lab Hangzhou China National Engineering Research Center for Big Data Technology and System Service Computing Technology and System Lab Cluster and Grid Computing Lab School of Software Engineering Huazhong University of Science and Technology Wuhan China Eastern Institute of Technology Ningbo China National University of Singapore Singapore Singapore Michigan Technological University Houghton United States School of Computer Science and Engieering UNSW Sydney Kensington Australia

Graph pattern mining is essential for deciphering complex networks. In the real world, graphs are dynamic and evolve over time, necessitating updates in mining patterns to reflect these changes. Traditional methods use fine-grained incremental computation to avoid full re-mining after each update, which improves speed but often overlooks potential gains from examining inter-update interactions holistically, thus missing out on overall efficiency *** this paper, we introduce Cheetah, a dynamic graph mining system that processes updates in a coarse-grained manner by leveraging exploration domains. These domains exploit the community structure of real-world graphs to uncover data reuse opportunities typically missed by existing approaches. Exploration domains, which encapsulate extensive portions of the graph relevant to updates, allow multiple updates to explore the same regions efficiently. Cheetah dynamically constructs these domains using a management module that identifies and maintains areas of redundancy as the graph changes. By grouping updates within these domains and employing a neighbor-centric expansion strategy, Cheetah minimizes redundant data accesses. Our evaluation of Cheetah across five real-world datasets shows it outperforms current leading systems by an average factor of 2.63 ×.

关键词： Graph mining system dynamic graph incremental computing

The Minority Matters: A Diversity-Promoting Collaborative Metric Learning Algorithm

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Bao, Shilong Xu, Qianqian Yang, Zhiyong He, Yuan Cao, Xiaochun Huang, Qingming State Key Laboratory of Information Security Institute of Information Engineering CAS China School of Cyber Security University of Chinese Academy of Sciences China Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS China School of Computer Science and Tech. University of Chinese Academy of Sciences China Alibaba Group China School of Cyber Science and Technology Shenzhen Campus Sun Yat-sen University China Key Laboratory of Big Data Mining and Knowledge Management CAS China Peng Cheng Laboratory China

Collaborative Metric Learning (CML) has recently emerged as a popular method in recommendation systems (RS), closing the gap between metric learning and Collaborative Filtering. Following the convention of RS, existing methods exploit unique user representation in their model design. This paper focuses on a challenging scenario where a user has multiple categories of interests. Under this setting, we argue that the unique user representation might induce preference bias, especially when the item category distribution is imbalanced. To address this issue, we propose a novel method called Diversity-Promoting Collaborative Metric Learning (DPCML), with the hope of considering the commonly ignored minority interest of the user. The key idea behind DPCML is to include a multiple set of representations for each user in the system. Based on this embedding paradigm, user preference toward an item is aggregated from different embeddings by taking the minimum item-user distance among the user embedding set. Furthermore, we observe that the diversity of the embeddings for the same user also plays an essential role in the model. To this end, we propose a diversity control regularization term to accommodate the multi-vector representation strategy better. Theoretically, we show that DPCML could generalize well to unseen test data by tackling the challenge of the annoying operation that comes from the minimum value. Experiments over a range of benchmark datasets speak to the efficacy of DPCML. © 2022, CC BY-NC-SA.

关键词： Embeddings

AMDET: Attention based Multiple Dimensions EEG Transformer for Emotion Recognition

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Xu, Yongling Du, Yang Zou, Jing Zhou, Tianying Xiao, Lushan Liu, Li Ma, Pengcheng The Big Data Center Nanfang Hospital Southern Medical University Guangzhou510515 China The Brainup Research Lab Naolu Technology Co. Ltd. Beijing100124 China The Department of Psychology and Economics University of California Los Angeles Los AngelesCA90024 United States The Department of Electrical Engineering and Computer Sciences UC Berkeley BerkeleyCA94720 United States The Department of Infectious Diseases Nanfang Hospital Southern Medical University Guangzhou510515 China

Affective computing is an important subfield of artificial intelligence, and with the rapid development of brain-computer interface technology, emotion recognition based on EEG signals has received broad attention. It is still a great challenge to effectively explore the multi-dimensional information in the EEG data in spite of a large number of deep learning methods . In this paper, we propose a deep model called Attention-based Multiple Dimensions EEG Transformer (AMDET), which can leverage the complementarity among the spectral-spatial-temporal features of EEG data by employing the multi-dimensional global attention mechanism. We transformed the original EEG data into 3D temporal-spectral-spatial representations and then the AMDET would use spectral-spatial transformer encoder layer to extract effective features in the EEG signal and concentrate on the critical time frame with a temporal attention layer. We conduct extensive experiments on the DEAP, SEED, and SEED-IV datasets to evaluate the performance of AMDET and the results outperform the state-of-the-art baseline on three datasets. Accuracy rates of 97.48%, 96.85%, 97.17%, 87.32% were achieved in the DEAP-Arousal, DEAP-Valence, SEED, and SEED-IV datasets, respectively. We also conduct extensive experiments to explore the possible brain regions that influence emotions and the coupling of EEG signals. Remarkably, AMDET can perform as well even with few channels which are identified by visualizing what the trained model learned. The accuracy could achieve over 90% even with only eight channels and it is of great use and benefit for practical applications. Copyright © 2022, The Authors. All rights reserved.

关键词： Emotion Recognition

Interpolation Estimation Method of Tropospheric Delay for Long Baseline Network RTK Based on Support Vector Machine

学校读者我要写书评

暂无评论

IOP Conference Series: Earth and Environmental Science 2018年第1期192卷

作者： Jian Deng Miaoqiang Xu Xuexiang Yu Aiguo Zhang School of Computer and Information Engineering Xia men University of Technology Xiamen 361024 China Big Data Institute of Digital Natural Disaster Monitoring in Fujian Xiamen University of Technology Xia men 361024 China Surveying and Mapping Institute Anhui University of Science and Technology Huainan 232000 China

In order to solve the problem of low precision of tropospheric delay interpolation under the long-distance sparse reference station, a method of tropospheric delay interpolation estimation based on support vector machine (SVM) theory was proposed. Firstly, the troposphere was assumed to be an infinitely thin single-layer membrane, and with the known information of reference stations, the puncture point coordinates and zenith tropospheric delay of each reference station on each visible satellite in the single-layer membrane were obtained. Then, the puncture point coordinates and zenith tropospheric delay were taken as training samples to optimize the appropriate core parameters of SVM and establish the SVM model of tropospheric delay. Finally, two sets of network RTK data with different lengths were selected to compare and analyze the effect of the interpolated tropospheric delay. The results showed that under the long-distance reference station, with the SVM tropospheric delay model established in this paper, the tropospheric delay accuracy of the interpolation estimation is better than 2cm, and the estimation error is generally stable, and the interpolation effect of satellite in different systems is basically the same.

关键词：

Artificial intelligence-assisted niacin skin flush screening in early psychosis identification and prediction

学校读者我要写书评

暂无评论

General Psychiatry 2022年第2期35卷 76-78页

作者： Tao Chen Haichun Liu Renfang Tian Ranpiao Gan Wenzuo Xu Tianhong Zhang Jijun Wang Department of Economics University of WaterlooWaterlooOntarioCanada Depatment of Statistics and Actuarial Science(Cross Appointed) University of Waterloo WaterlooOntarioCanada Big Data Research Lab Universty of WaterlooWaterlooOntarioCanada Shanghai Niacin Tech Co. LtdShanghal China Labor and Worklife Program Harvard UniversityCambridgeMassachusettsUSA Department of Automation Shanghai Jao Tong UniversityShanghaiChina Ningbo Artificial Intelligent Institute Shanghai Jiao Tong UniversityNingboChina School of Management Economicsand MathematicsKing's University CollegeLondon0ntarioCanada Shanghai Intelligent Psychological Evaluaton and Intervention Engineering Technology Research Center Shanghai Key Laboratory of Psychotic DisordersShanghai Mental Health CenterShanghai Jiao Tong University School of MedicineShanghaiChina Bio-X Institutes Key Laboratory for the Genetics of Developmental and Neuropsychiatric Disorders(Ministry of Education)ShanghaiChina CAS Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of SciencesShanghaiChina Institute of Psychology and Behavioral Science Shanghai Jiao Tong UniversiyShanghaiChina

Schizophrenia is a devastating mental disorder affecting 20 million people *** diagnosis is crucial for disease management and improvement in prognosis,and diagnostic biomarkerscan serveasobjective indicators for the early screening of the *** on the observation of diminished flush responses to niacin in patients with schizophrenia Horrobin proposed anoninvasive niacin skin flush screening for schizophrenia.

关键词： diagnosis prognosis prediction

Taking the pulse of COVID-19:a spatiotemporal perspective

学校读者我要写书评

暂无评论

International Journal of Digital Earth 2020年第10期13卷 1186-1211页

作者： Chaowei Yang Dexuan Sha Qian Liu Yun Li Hai Lan Weihe Wendy Guan Tao Hu Zhenlong Li Zhiran Zhang John Hoot Thompson Zifu Wang David Wong Shiyang Ruan Manzhu Yu Douglas Richardson Luyao Zhang Ruizhi Hou You Zhoua Cheng Zhong Yifei Tian Fayez Beaini Kyla Carte Colin Flynn Wei Liu Dieter Pfoser Shuming Bao Mei Li Haoyuan Zhang Chunbo Liu Jie Jiang Shihong Du Liang Zhao Mingyue Lu Lin Li Huan Zhou Andrew Ding NSF Spatiotemporal Innovation Center George Mason Univ.FairfaxVAUSA Department of Geography and GeoInformation Science George Mason Univ.FairfaxVAUSA Department of Geographical Sciences University of MarylandCollege ParkMDUSA Center for Geographic Analysis Harvard UniversityCambridgeMAUSA Geoinformation and Big Data Research Lab Department of GeographyUniversity of South CarolinaColumbiaSCUSA Chinese Academy of Surveying and Mapping BeijingPeople’s Republic of China School of Resource and Environmental Sciences Wuhan UniversityWuhanPeople’s Republic of China Cloud Computeand Storage OperationsInformation Technology ServicesGeorge Mason UniversityFairfaxVAUSA Department of Geography Pennsylvania State UniversityState CollegePAUSA School of Business Management East China Normal UniversityShanghaiPeople’s Republic of China School of Mathematical Sciences East China Normal UniversityShanghaiPeople’s Republic of China Department of Psychology George Mason UniversityFairfaxVAUSA mDepartment of Computer Science George Mason UniversityFairfaxVAUSA Department of Information Technology George Mason UniversityFairfaxVAUSA Department of Biology George Mason UniversityFairfaxVAUSA College of Land Science and Technology China Agricultural UniversityBeijingPeople’s Republic of China China Data Institute Ann ArborMIUSA Institute of Remote Sensing and GIS Peking UniversityBeijingPeople’s Republic of China Aerospace Information Research Institute Chinese Academy of SciencesBeijingPeople’s Republic of China School of Geomatics and Urban Spatial Informatics Beijing University of Civil Engineering and ArchitectureBeijingPeople’s Republic of China School of Geographical Sciences Nanjing University of Information Science&TechnologyNanjingPeople’s Republic of China School of Geodesy and Geomatics Wuhan UniversityWuhanPeople’s Republic of China wFaculty of Environment University of WaterlooWaterlooCanada

The sudden outbreak of the Coronavirus disease(COVID-19)swept across the world in early 2020,triggering the lockdowns of several billion people across many countries,including China,Spain,India,the U.K.,Italy,France,Germany,Brazil,Russia,and the *** transmission of the virus accelerated rapidly with the most confirmed cases in the U.S.,India,Russia,and *** response to this national and global emergency,the NSF Spatiotemporal Innovation Center brought together a taskforce of international researchers and assembled implementation strategies to rapidly respond to this crisis,for supporting research,saving lives,and protecting the health of global *** perspective paper presents our collective view on the global health emergency and our effort in collecting,analyzing,and sharing relevant data on global policy and government responses,human mobility,environmental impact,socioeconomical impact;in developing research capabilities and mitigation measures with global scientists,promoting collaborative research on outbreak dynamics,and reflecting on the dynamic responses from human societies.

关键词： big data Earth system emergency geospatial sciences epidemics applications