检索结果-内蒙古大学图书馆

Sindiff: Spoken-to-Sign Language Generation Based Transformer Diffusion Model

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Liang, Wuyan Xu, Xiaolong Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Nanjing University of Posts and Telecommunications No. 9 Wenyuan Road Nanjing210023 China School of Computer Science Nanjing University of Posts and Telecommunications No. 9 Wenyuan Road Nanjing210023 China

Sign language generation (SLG) is meaningful in facilitating the communication between deaf and healthy people. Recently, transformer-based diffusion models have been achieved success in human motion generation. However, it is still under-explored how to generate long-term sign language with diffusion model. Here, we propose a transformer-based diffusion framework, SinDiff, that explore all the advantages of transformer (i.e., dynamic attention and global context) for spoken-driven SLG. To meet the goal, we develop a linear biased multi-head self-attention with periodic position encoding to generalize longer-term sign language, and a linear biased cross-modal multi-head attention function to align the spoken and sign language modalities. Besides, a transformer-based spoken encoder is introduced to solve the limitations of sign data and learn the global spoken semantics. By experimental results on our testbed, we show that the word accuracy of the SinDiff for the back-translated spoken translation with ground truth spoken reach 21.02%, which has a 2.8% improvement over the 18.22% accuracy of the typical SLG method (GEN-OBT) on PHOENIX14T test set. And the generative visualization results further verify the feasibility of the proposed method. © 2023, The Authors. All rights reserved.

关键词： Semantics

Optimizing Location Quality in Privacy Preserving Crowdsensing

学校读者我要写书评

暂无评论

Optimizing Location Quality in Privacy Preserving Crowdsensi...

2019 IEEE Global Communications Conference (GLOBECOM)

作者： Yuhui Zhang Ming Li Dejun Yang Jian Tang Guoliang Xue Colorado School of Mines Golden CO Colorado School of Mines Golden CO USA Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing University of Posts and Telecommunications Nanjing China Syracuse University Syracuse NY Arizona State University Tempe AZ

ISBN: (数字)9781728109626

ISBN: (纸本)9781728109633

Crowdsensing enables a wide range of data collection, where the data are usually tagged with private locations. Protecting users' location privacy has been a central issue. The study of various location perturbation techniques for protecting users' location privacy has received widespread attention. Despite the huge promise and considerable attention, the location perturbation operation causes inevitable location errors, which can diminish the location quality of the crowdsensing results. Provable good algorithms that consider location quality in privacy preserving crowdsensing from optimization perspectives are still lacking in the literature. In this paper, we investigate the problem of location quality optimization in privacy preserving crowdsensing, which is to minimize the location quality desegregation, while protecting all users' location privacy. We present an optimal algorithm OLQDM for this problem. Extensive simulations demonstrate that OLQDM significantly outperforms an existing algorithm in terms of the location quality and SSE.

关键词： Privacy Degradation Sensors Perturbation methods Optimization Minimization Mobile handsets

Consistent and Complementary Graph Regularized Multi-view Subspace Clustering

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zheng, Qinghai Zhu, Jihua Li, Zhongyu Pang, Shanmin Wang, Jun Chen, Lei School of Software Engineering Xi’an Jiaotong University Xi’an710049 China Shanghai Institute for Advanced Communication and Data Science School of Communication and Information Engineering Shanghai University Shanghai200444 China Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing University of Posts and Telecommunications Nanjing210023 China

This study investigates the problem of multi-view clustering, where multiple views contain consistent information and each view also includes complementary information. Exploration of all information is crucial for good multi-view clustering. However, most traditional methods blindly or crudely combine multiple views for clustering and are unable to fully exploit the valuable information. Therefore, we propose a method that involves consistent and complementary graph-regularized multi-view subspace clustering (GRMSC), which simultaneously integrates a consistent graph regularizer with a complementary graph regularizer into the objective function. In particular, the consistent graph regularizer learns the intrinsic affinity relationship of data points shared by all views. The complementary graph regularizer investigates the specific information of multiple views. It is noteworthy that the consistent and complementary regularizers are formulated by two different graphs constructed from the first-order proximity and second-order proximity of multiple views, respectively. The objective function is optimized by the augmented Lagrangian multiplier method in order to achieve multi-view clustering. Extensive experiments on six benchmark datasets serve to validate the effectiveness of the proposed method over other state-of-the-art multi-view clustering methods. Copyright © 2020, The Authors. All rights reserved.

关键词： Clustering algorithms

A Multimodal Ecological Civilization Pattern Recommendation Method Based on Large Language Models and Knowledge Graph

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yu, Zhihang Wang, Shu Zhu, Yunqiang Zou, Zhiqiang College of Computer Nanjing University of Posts and Telecommunications Nanjing210023 China State Key Laboratory of Resources and Environmental Information System Institute of Geographic Sciences and Natural Resources Research Chinese Academy of Sciences Beijing China Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Nanjing University of Posts and Telecommunications Nanjing210023 China

The Ecological Civilization Pattern Recommendation System (ECPRS) aims to recommend suitable ecological civilization patterns for target regions, promoting sustainable development and reducing regional disparities. However, the current representative recommendation methods are not suitable for recommending ecological civilization patterns in a geographical context. There are two reasons for this. Firstly, regions have spatial heterogeneity, and the (ECPRS)needs to consider factors like climate, topography, vegetation, etc., to recommend civilization patterns adapted to specific ecological environments, ensuring the feasibility and practicality of the recommendations. Secondly, the abstract features of the ecological civilization patterns in the real world have not been fully utilized., resulting in poor richness in their embedding representations and consequently, lower performance of the recommendation system. Considering these limitations, we propose the ECPR-MML method. Initially, based on the novel method UGPIG, we construct a knowledge graph to extract regional representations incorporating spatial heterogeneity features. Following that, inspired by the significant progress made by Large Language Models (LLMs) in the field of Natural Language processing (NLP), we employ Large LLMs to generate multimodal features for ecological civilization patterns in the form of text and images. We extract and integrate these multimodal features to obtain semantically rich representations of ecological civilization. Through extensive experiments, we validate the performance of our ECPR-MML model. Our results show that F1@5 is 2.11% higher compared to state-of-the-art models, 2.02% higher than NGCF, and 1.16% higher than UGPIG. Furthermore, multimodal data can indeed enhance recommendation performance. However, the data generated by LLM is not as effective as real data to a certain extent." Copyright © 2023, The Authors. All rights reserved.

关键词： Recommender systems

PhotoRedshift-MML: a multimodal machine learning method for estimating photometric redshifts of quasars

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Hong, Shuxin Zou, Zhiqiang Luo, A-Li Kong, Xiao Yang, Wenyu Chen, Yanli College of Computer Nanjing University of Posts and Telecommunications Nanjing210023 China Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing210023 China CAS Key Laboratory of Optical Astronomy National Astronomical Observatories Beijing100101 China School of Astronomy and Space Science University of Chinese Academy of Sciences Beijing100049 China

We propose a Multimodal Machine Learning method for estimating the Photometric Redshifts of quasars (PhotoRedshift-MML for short), which has long been the subject of many investigations. Our method includes two main models, i.e. the feature transformation model by multimodal representation learning, and the photometric redshift estimation model by multimodal transfer learning. The prediction accuracy of the photometric redshift was significantly improved owing to the large amount of information offered by the generated spectral features learned from photometric data via the MML. A total of 415,930 quasars from Sloan digital Sky Survey (SDSS) data Release 17, with redshifts between 1 and 5, were screened for our experiments. We used |Δz| = |(zphot− zspec)/(1 + zspec)| to evaluate the redshift prediction and demonstrated a 4.04% increase in accuracy. With the help of the generated spectral features, the proportion of data with |Δz| Copyright © 2022, The Authors. All rights reserved.

关键词： Surveys

Adaptive Teacher-and-Student Model for Heterogeneous Domain Adaptation

学校读者我要写书评

暂无评论

Adaptive Teacher-and-Student Model for Heterogeneous Domain ...

IEEE International Conference on data Mining (ICDM)

作者： Yunyun Wang Xuzhang Chen Songcan Chen Hui Xue College of Computer Science Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Nanjing University of Posts & Telecommunications Nanjing China College of Computer Science & Technology/AI MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Nanjing University of Aeronautics & Astronautics Nanjing China College of Computer Science and Engineering Southeast University Nanjing China

In heterogeneous domain adaptation (HDA), since the feature spaces of the source and target domains are different, knowledge transfer from the source to the target domain is really challenging. How to align the different feature spaces and then adaptively transfer the related knowledge is critical for HDA. In this paper, we develop an adaptive teacher-and-student model for heterogeneous domain adaptation (AtsHDA). In AtsHDA, the source domain as a teacher and the target domain as a student are aligned or co-adapted to each other first, so that their correlation can be maximized. Then the target domain adaptively learns from the source domain. Specifically, there is a balance between the learning by the target domain itself and the instruction from the source domain. That is, when the guidance from the source domain is helpful for learning, the learning of target classifier emphasizes the instruction of source knowledge, and considers its own knowledge more, otherwise. Further, an ensemble method is designed to decide such a balance. Finally, empirical results show that AtsHDA can achieve competitive results compared with the state-of-arts.

关键词：

Network-coding-based video transmission via WiFi direct in local mobile networks

学校读者我要写书评

暂无评论

Network-coding-based video transmission via WiFi direct in l...

International Conference on intelligent System and Knowledge Engineering, ISKE

作者： Lei Wang Qing Wang Zhongwei Sun Jun Yin Lijie Xu Yuwang Yang Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing P.R. China School of Computer Nanjing University of Posts and Telecommunications Nanjing P.R. China School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing P.R. China

It will incur high expense by sharing large-scale files (such as video) with others through commercial 4G networks. When the users are located at the same area, the traffic cost could be saved via the technology of WiFi direct. In this paper, a network-coding-based video transmission scheme is proposed for such scenario. In the scheme, when a source user needs to transmit a video to others, it equally splits the original file into multiple segments. When a neighbour connects to it and requests a segment, the user will linearly generate a new segment with all the segments it buffers, and then send the re-encoded segment to other neighbours. After the neighbours receive all or part of the segments, they could become a new source and send data to their own neighbours through the technology of WiFi direct. The results show that compared with the traditional scheme, the transmission scheme based on network coding could be used to complete transmission at a higher rate.

关键词： Switches Analytical models Network coding Streaming media Wireless fidelity Mobile handsets Mobile computing

VFL: A Verifiable Federated Learning with Privacy-Preserving for Big data in Industrial IoT

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Fu, Anmin Zhang, Xianglong Xiong, Naixue Gao, Yansong Wang, Huaqun School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210094 China Northeastern State University Department of Mathematics and Computer Science TahlequahOK United States Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing University of Posts and Telecommunications Nanjing210023 China

Due to the strong analytical ability of big data, deep learning has been widely applied to model the collected data in industrial IoT. However, for privacy issues, traditional data-gathering centralized learning is not applicable to industrial scenarios sensitive to training sets. Recently, federated learning has received widespread attention, since it trains a model by only relying on gradient aggregation without accessing training sets. But existing researches reveal that the shared gradient still retains the sensitive information of the training set. Even worse, a malicious aggregation server may return forged aggregated gradients. In this paper, we propose the VFL, verifiable federated learning with privacy-preserving for big data in industrial IoT. Specifically, we use Lagrange interpolation to elaborately set interpolation points for verifying the correctness of the aggregated gradients. Compared with existing schemes, the verification overhead of VFL remains constant regardless of the number of participants. Moreover, we employ the blinding technology to protect the privacy of the gradients submitted by the participants. If no more than n-2 of n participants collude with the aggregation server, VFL could guarantee the encrypted gradients of other participants not being inverted. Experimental evaluations corroborate the practical performance of the presented VFL framework with high accuracy and efficiency. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep learning

A Point Cloud Completion Method for Industrial Component Based on Multiple Encoders

学校读者我要写书评

暂无评论

A Point Cloud Completion Method for Industrial Component Bas...

Chinese Automation Congress (CAC)

作者： Lei Wang Ran Chen Wei Huang Yiping Chen Jun Yin School of Computer Nanjing University of Posts and Telecommunications Nanjing P. R. China Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing P. R. China School of Internet of Things Nanjing University of Posts and Telecommunications Nanjing P. R. China

In real-world scenarios, laser-scanned point clouds often suffer from missing data points due to occlusion and other factors, significantly impacting subsequent tasks. We propose a novel point cloud completion method based on self-attention mechanisms and multi-layer perceptrons to address this issue, building upon the PCN model. Our experimental evaluations on the PCN dataset demonstrate notable enhancements in the completion effectiveness and evaluation metrics compared to the baseline model. Additionally, focusing on the point cloud completion context for industrial components, we introduce an approach for constructing an industrial component point cloud dataset. This method leverages 3D mesh models to swiftly generate complete point clouds with uniform distribution and devoid of noise while simulating the incomplete point clouds obtained from real-world scanning scenarios. The final results fully exhibit the feasibility of this scheme, which is a strong candidate for future industrial component point cloud completion.

关键词：