检索结果-内蒙古大学图书馆

Adaptive Loose Optimization for Robust Question Answering

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ma, Jie Wang, Pinghui Wang, Zewei Kong, Dechen Hu, Min Han, Ting Liu, Jun The Ministry of Education of Key Laboratory for Intelligent Networks and Network Security School of Cyber Science and Engineering Xi’an Jiaotong University Shaanxi Xi’an710049 China The Ministry of Education of Key Laboratory for Intelligent Networks and Network Security School of Automation Science and Engineering Xi’an Jiaotong University Shaanxi Xi’an710049 China The China Mobile Research Institute China The Shannxi Provincial Key Laboratory of Big Data Knowledge Engineering School of Computer Science and Technology Xi’an Jiaotong University Shaanxi Xi’an710049 China

Question answering methods are well-known for leveraging data bias, such as the language prior in visual question answering and the position bias in machine reading comprehension (extractive question answering). Current debiasing methods often come at the cost of significant in-distribution performance to achieve favorable out-of-distribution generalization ability, while non-debiasing methods sacrifice a considerable amount of out-of-distribution performance in order to obtain high in-distribution performance. Therefore, it is challenging for them to deal with the complicated changing real-world situations. In this paper, we propose a simple yet effective novel loss function with adaptive loose optimization, which seeks to make the best of both worlds for question answering. Our main technical contribution is to reduce the loss adaptively according to the ratio between the previous and the current optimization state on mini-batch training data. This loose optimization can be used to prevent non-debiasing methods from overlearning data bias while enabling debiasing methods to maintain slight bias learning. Experiments on the visual question answering datasets, including VQA v2, VQA-CP v1, VQA-CP v2, GQA-OOD, and the extractive question answering dataset SQuAD demonstrate that our approach enables QA methods to obtain state-of-the-art in- and out-of-distribution performance in most cases. The source code has been released publicly in https://***/reml-group/ALO. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

Space-Air-Ground Integrated network (SAGIN) for 6G: Requirements, Architecture and Challenges

学校读者我要写书评

暂无评论

China Communications 2022年第2期19卷 90-108页

作者： Huanxi Cui Jun Zhang Yuhui Geng Zhenyu Xiao Tao Sun Ning Zhang Jiajia Liu Qihui Wu Xianbin Cao School of Electronic and Information Engineering Beijing Key Laboratory for Network-Based Cooperative Air application ManagementBeihang UniversityBeijing 100191China Advanced Research Institute of Multidisciplinary Science Beijing Institute of TechnologyBeijing 100081China China Mobile Research Institute Beijing 100053China Department of Electrical and Computer Engineering University of WindsorONCanadaN9B 3P4 National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology the School of CybersecurityNorthwestern Polytechnical UniversityXi’anShaanxi710072China College of Electronics and Information Engineering Nanjing University of Aeronautics and AstronauticsNanjing 210007China

As the fifth-generation(5G)mobile communication network may not meet the requirements of emerging technologies and applications,including ubiquitous coverage,industrial internet of things(IIoT),ubiquitous artificial intelligence(AI),digital twins(DT),etc.,this paper aims to explore a novel space-air-ground integrated network(SAGIN)architecture to support these new requirements for the sixth-generation(6G)mobile communication network in a flexible,low-latency and efficient ***,we first review the evolution of the mobile communication network,followed by the application and technology requirements of *** the current 5G non-terrestrial network(NTN)architecture in supporting the new requirements is deeply *** that,we proposes a new flexible,low-latency and flat SAGIN architecture,and presents corresponding use ***,the future research directions are discussed.

关键词： 6G AI DT SAGIN NTN network architecture

XTQA: Span-Level Explanations for Textbook Question Answering

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Ma, Jie Chai, Qi Liu, Jun Yin, Qingyu Wang, Pinghui Zheng, Qinghua Ministry of Education of Key Lab for Intelligent Networks and Network Security School of Cyber Science and Engineering Xi’an Jiaotong University Shaanxi Xi’an710049 China Shaanxi Provincial Key Laboratory of Big Data Knowledge Engineering The School of Computer Science and Technology Xi’an Jiaotong University Shaanxi Xi’an710049 China Amazon United States

Textbook Question Answering (TQA) is the task of correctly answering diagram or non-diagram questions given large multi-modal contexts consisting of abundant essays and diagrams. In real-world scenarios, an explainable TQA system plays a key role in deepening humans’ understanding of learned knowledge. However, there is no work to investigate how to provide explanations currently. To address this issue, we devise a novel architecture towards span-level eXplanations for TQA (XTQA). In this paper, spans are the combinations of sentences within a paragraph. The key idea is to consider the entire textual context of a lesson as candidate evidence, and then use our proposed coarse-to-fine grained Explanation Extracting (EE) algorithm to narrow down the evidence scope and extract the span-level explanations with varying lengths for answering different questions. The EE algorithm can also be integrated into other TQA methods to make them explainable and improve the TQA performance. Experimental results show that XTQA obtains the best overall explanation result (mIoU) of 52.38% on the first 300 questions of CK12-QA test splits, demonstrating the explainability of our method (non-diagram: 150 and diagram: 150). The results also show that XTQA achieves the best TQA performance of 36.46% and 36.95% on the aforementioned splits respectively. We have released our code in https://***/dr-majie/*** Codes 68T07 Copyright © 2020, The Authors. All rights reserved.

关键词： Textbooks

Learnable Convolutional Attention network for Knowledge Graph Completion

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Shang, Bin Zhao, Yinliang Liu, Jun School of Computer Science and Technology Xi’an Jiaotong University Shaanxi Xi’an710049 China Shaanxi Province Key Laboratory of Satellite and Terrestrial Network Tech. R&D Xi’an Jiaotong University Shaanxi Xi’an710049 China National Engineering Lab for Big Data Analytics Shaanxi Xi’an710049 China

Recently, graph convolutional networks (GCNs) and graph attention networks (GATs) have been widely used in knowledge graph completion (KGC), which aims to solve the incompleteness of knowledge graphs (KGs). However, GCNs based KGC models tend to stack redundant information when one entity have several noisy neighbors. GATs based KGC models may focus on certain neighbors and lead to weakening of structural information. Furthermore, relations in KGs have their specific semantics and should be considered when aggregating neighbor information (message passing). To address the above limitations, we propose a learnable convolutional attention network for knowledge graph completion named LCA-KGC. LCA-KGC introduces a knowledge graph convolutional attention network by using the convolution operation before the attention mechanism to ensure the structural information acquisition and avoid the redundant information stacking. Furthermore, to complete the autonomous switching of GNNs types and eliminate the necessity of pre-validating the local structure of KGs, LCA-KGC designs a learnable knowledge graph convolutional attention network by comprising three types of GNNs in one learnable formulation. Moreover, a learnable message function is proposed to emphasize relational semantics when aggregating neighbor information. Extensive experiments on standard KG datasets validate the effectiveness of the proposed innovations, and LCA-KGC achieves state-of-the-art (SOTA) performance compared to existing approaches (e.g., compared to SOTA approaches, LCA-KGC improves MRR from 0.360 to 0.372 on FB15k-237 dataset, and Hits@3 from 0.561 to 0.581 on YAGO3-10 dataset). © 2023, The Authors. All rights reserved.

关键词： Graph neural networks

Visual Tracking Algorithm Based on Color Name Histogram

学校读者我要写书评

暂无评论

Visual Tracking Algorithm Based on Color Name Histogram

作者： Chenchen Yue Wangsheng Yu Zhiqiang Hou Sugang Ma College of Computer Science and Technology Xi'an University of Posts and Telecommunications Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'an University of Posts and Telecommunications Information and Navigation Institute of Air Force Engineering University

Aimed at the problem that traditional histogram is sensitive to illumination changes in visual tracking,combined with the CN(Color Name) feature,we proposed a new feature(denotes CNH,Color Name Histogram) based on color ***,the method projected the original RGB image to CN space to obtain robust 11 feature ***,we counted the each pixel numbers of feature ***,normalizing the amount of pixels in each *** addition,we adopted a feature adaptive fusion method to combine CNH and HOG(Histogram of Oriented Gradient).In order to prove validity of the proposed algorithm,we use Staple(Sum of Template And Pixel-wise Learners) algorithm frame to make a controlled *** contrast with the reference algorithms,the success of our algorithm increases by 1.5% and the precision increases by 1.7%.The results show that this method retains the advantages of traditional histogram which is insensitive to target deformation,but also enhances the robustness to illumination change.

关键词： Visual tracking Color histogram Color name Feature fusion

Hypertensive Pulse Diagnosis Method Based on Hilbert-Huang Transform and Feature Fusion Dimensionality Reduction 22

学校读者我要写书评

暂无评论

Hypertensive Pulse Diagnosis Method Based on Hilbert-Huang T...

Proceedings of the 2022 5th International Conference on Artificial Intelligence and Pattern Recognition

作者： Lin Fan Ruiling Yao Zhongmin Wang School of Computer Science and Technology Xi'an University of Posts and Telecommunications China and Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing China School of Computer Science and Technology Xi'an University of Posts and Telecommunications China

ISBN: (纸本)9781450396899

With the development of artificial intelligence, pulse diagnosis has been standardized and objectified. However, there is a lack of research on the extraction and dimensionality reduction of hypertensive pulse features. We propose two effective features for distinguishing pulses of disease samples and a fusion dimensionality reduction method that combines linear and nonlinear dimensionality reduction. The results show that the proposed features and dimensionality reduction method make the classification accuracy of hypertension pulse feature reach 94.23%, and the training time of the classifier is reduced by 47 seconds, which improves the performance in terms of both accuracy and time.

关键词： Hilbert-Huang Transform Hypertension Principal Component Analysis Pulse Diagnosis

Backdoor Defense with Machine Unlearning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Liu, Yang Fan, Mingyuan Chen, Cen Liu, Ximeng Ma, Zhuo Wang, Li Ma, Jianfeng Xi'An China Shaanxi Key Laboratory of Network and System Security Xidian University Xi'An China College of Computer and Data Science Fuzhou University Fuzhou China School of Data Science and Engineering East China Normal University Shanghai China Ant Group Hangzhou China

Backdoor injection attack is an emerging threat to the security of neural networks, however, there still exist limited effective defense methods against the attack. In this paper, we propose BAERASER, a novel method that can erase the backdoor injected into the victim model through machine unlearning. Specifically, BAERASER mainly implements backdoor defense in two key steps. First, trigger pattern recovery is conducted to extract the trigger patterns infected by the victim model. Here, the trigger pattern recovery problem is equivalent to the one of extracting an unknown noise distribution from the victim model, which can be easily resolved by the entropy maximization based generative model. Subsequently, BAERASER leverages these recovered trigger patterns to reverse the backdoor injection procedure and induce the victim model to erase the polluted memories through a newly designed gradient ascent based machine unlearning method. Compared with the previous machine unlearning solutions, the proposed approach gets rid of the reliance on the full access to training data for retraining and shows higher effectiveness on backdoor erasing than existing fine-tuning or pruning methods. Moreover, experiments show that BAERASER can averagely lower the attack success rates of three kinds of state-of-the-art backdoor attacks by 99% on four benchmark datasets. © 2022, CC BY-NC-ND.

关键词： network security

A High Parallel Accelerator Design Based on the Sparsity of Convolutional Neural network

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Pan, Xiao Ying Mu, Yuanzhen Jia, Ningxin Gao, Xuan Rong Feng, Tong School of Computer Science and Technology Xi’an University of Posts and Telecommunications Shannxi Xi’an710121 China Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi’an University of Posts and Telecommunications Shannxi Xi’an710121 China

Designing a domain-specific accelerator is an effective way to improve the performance of inference network on resource-constrained embedded devices. However, there are a large number of invalid memory accesses and calculations in the convolutional neural network. Based on this, this paper proposes an accelerator that adopts a flexible neural network structure that adjusts the processing flow of the network, reduces the size of the feature map sent into the activation function, and greatly reduces the inherent invalid calculation and memory access of the network. The accelerator uses loop expansion and loop reconstruction to solve the read-write problem in convolution operation, realize instruction-level parallelism and data-level parallelism. We determine the design parameters of parallel computing by modelling the constrained single-objective optimization problem to find the optimal cyclic expansion factor. Evaluate the performance, resource utilization, and energy consumption of accelerators in different neural networks. Compared with the general processor CPU and GPU, the running time of the accelerator is shortened by 6.89s and 1.16s, and the power consumption is only 7.47% of CPU and 3% of GPU. Compared with the state-of-the-art accelerator, the performance is increased by 2.8x and the power consumption is reduced by approximately 6W. © 2022, The Authors. All rights reserved.

关键词： Field programmable gate arrays (FPGA)

A Low-rank Matching Attention based Cross-modal Feature Fusion Method for Conversational Emotion Recognition

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Shou, Yuntao Liu, Huan Cao, Xiangyong Meng, Deyu Dong, Bo School of Computer Science and Technology The Ministry of Education Key Laboratory for Intelligent Networks and Network Security The Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi’an Jiaotong Shaanxi Xi’an710049 China School of Computer Science and Technology Xi’an Jiaotong Shaanxi Xi’an710049 China The BigKE Joint Innovation Center SHMEEA Shanghai China School of Mathematics and Statistics Xi’an Jiaotong University Xi’an China

Conversational emotion recognition (CER) is an important research topic in human-computer interactions. Although recent advancements in transformer-based cross-modal fusion methods have shown promise in CER tasks, they tend to overlook the crucial intra-modal and inter-modal emotional interaction or suffer from high computational complexity. To address this, we introduce a novel and lightweight cross-modal feature fusion method called Low-Rank Matching Attention Method (LMAM). LMAM effectively captures contextual emotional semantic information in conversations while mitigating the quadratic complexity issue caused by the self-attention mechanism. Specifically, by setting a matching weight and calculating inter-modal features attention scores row by row, LMAM requires only one-third of the parameters of self-attention methods. We also employ the low-rank decomposition method on the weights to further reduce the number of parameters in LMAM. As a result, LMAM offers a lightweight model while avoiding overfitting problems caused by a large number of parameters. Moreover, LMAM is able to fully exploit the intra-modal emotional contextual information within each modality and integrates complementary emotional semantic information across modalities by computing and fusing similarities of intra-modal and inter-modal features simultaneously. Experimental results verify the superiority of LMAM compared with other popular cross-modal fusion methods on the premise of being more lightweight. Also, LMAM can be embedded into any existing state-of-the-art CER methods in a plug-and-play manner, and can be applied to other multi-modal recognition tasks, e.g., session recommendation and humour detection, demonstrating its remarkable generalization ability. Copyright © 2023, The Authors. All rights reserved.

关键词： Semantics