检索结果-内蒙古大学图书馆

Securely Fine-tuning Pre-trained Encoders Against Adversarial Examples

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhou, Ziqi Li, Minghui Liu, Wei Hu, Shengshan Zhang, Yechao Wan, Wei Xue, Lulu Zhang, Leo Yu Yao, Dezhong Jin, Hai National Engineering Research Center for Big Data Technology and System China Services Computing Technology and System Lab. China Cluster and Grid Computing Lab. China Hubei Engineering Research Center on Big Data Security China Hubei Key Laboratory of Distributed System Security China School of Computer Science and Technology Huazhong University of Science and Technology China School of Software Engineering Huazhong University of Science and Technology China School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Information and Communication Technology Griffith University Australia

With the evolution of self-supervised learning, the pre-training paradigm has emerged as a predominant solution within the deep learning landscape. Model providers furnish pre-trained encoders designed to function as versatile feature extractors, enabling downstream users to harness the benefits of expansive models with minimal effort through fine-tuning. Nevertheless, recent works have exposed a vulnerability in pre-trained encoders, highlighting their susceptibility to downstream-agnostic adversarial examples (DAEs) meticulously crafted by attackers. The lingering question pertains to the feasibility of fortifying the robustness of downstream models against DAEs, particularly in scenarios where the pre-trained encoders are publicly accessible to the attackers. In this paper, we initially delve into existing defensive mechanisms against adversarial examples within the pre-training paradigm. Our findings reveal that the failure of current defenses stems from the domain shift between pre-training data and downstream tasks, as well as the sensitivity of encoder parameters. In response to these challenges, we propose Genetic Evolution-Nurtured Adversarial Fine-tuning (Gen-AF), a two-stage adversarial fine-tuning approach aimed at enhancing the robustness of downstream models. Gen-AF employs a genetic-directed dual-track adversarial fine-tuning strategy in its first stage to effectively inherit the pre-trained encoder. This involves optimizing the pre-trained encoder and classifier separately while incorporating genetic regularization to preserve the model’s topology. In the second stage, Gen-AF assesses the robust sensitivity of each layer and creates a dictionary, based on which the top-k robust redundant layers are selected with the remaining layers held fixed. Upon this foundation, we conduct evolutionary adaptability fine-tuning to further enhance the model’s generalizability. Our extensive experiments, conducted across ten self-supervised training methods and six

关键词： Topology

A Method of Realizing External Network Access to Intranet for Embedded Web Server 2

学校读者我要写书评

暂无评论

A Method of Realizing External Network Access to Intranet fo...

2nd International Conference on Electronic Information Technology and Smart Agriculture, ICEITSA 2022

作者： Zhang, Xian Qu, Taiguo Tang, Pengju Liu, Yiwen Key Lab. of Intelligent Contr. Technol. for Wuling Mountain Ecological Agriculture in Hunan Province Huaihua China Mountain Health Big Data Intelligent Processing and Application in Hunan Province Universities Key Laboratory of Wuling Huaihua China College of Computer Science and Electronic Engineering Hunan University ChangSha China School of Computer Science and Engineering Huaihua University Huaihua China

ISBN: (纸本)9781665464017

Based on the application requirements of the embedded Web server for external network access to the internal network, this paper designs and builds an embedded Web server scheme for external network access to the internal network based on the embedded Boa server and phddns program. This solution mainly transplants the Boa server and phddns program in embedded devices. Boa provides a Web server application function, and phddns software realizes the mapping and binding mapping between the IP address and the domain name of embedded devices in the intranet. An experiment is designed to test this scheme's connection success rate, response time, stable running time, domain name mapping success rate, CPU utilization and other performance indicators. The test results show that this scheme can better meet the system requirements of external network access to the internal network. Therefore, this scheme can support the information and intelligent monitoring and control of embedded devices and has good practical value and application prospects. © 2022 IEEE.

关键词： Intranets

SP3: Enhancing Structured Pruning via PCA Projection

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Hu, Yuxuan Zhang, Jing Zhao, Zhe Zhao, Chen Chen, Xiaodong Li, Cuiping Chen, Hong School of Information Renmin University of China Beijing China Key Laboratory of Data Engineering and Knowledge Engineering MOE China Engineering Research Center of Database and Business Intelligence MOE China Tencent AI Lab Tencent Beijing China School of Computer Science and Technology Xi'an Jiaotong University Xi'An China

Structured pruning is a widely used technique for reducing the size of pre-trained language models (PLMs), but current methods often overlook the potential of compressing the hidden dimension (d) in PLMs, a dimension critical to model size and efficiency. This paper introduces a novel structured pruning approach, Structured Pruning with PCA Projection (SP3), targeting the effective reduction of d by projecting features into a space defined by principal components before masking. Extensive experiments on benchmarks (GLUE and SQuAD) show that SP3 can reduce d by 70%, compress 94% of the BERTbase model, and maintain over 96% accuracy and outperform other methods that compress d by 6% in accuracy at the same compression ratio. SP3 has also proven effective with other models, including OPT and Llama. Our data and code are availab.e at ours repo. © 2023, CC BY.

关键词：

Global Green Digital Economy: Challenges and Development: Preface for ITQM 2022

学校读者我要写书评

暂无评论

Procedia Computer Science 2022年第C期214卷 1-3页

作者： Shi, Yong Li, Jianping Kou, Gang Berg, Daniel Tien, James College of Information Science and Technology University of Nebraska at Omaha OmahaNE68182 United States Research Centre on Fictitious Economy and Data Science Chinese Academy of Sciences Beijing100190 China Key Lab of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing100190 China School of Economics and Management University of Chinese Academy of Sciences Beijing100190 China School of Business Administration Southwestern University of Finance and Economics Chengdu611130 China College of Engineering University of Miami Coral GablesFL33124 United States

Development of Pillbox-Distributed-Fed Antenna Arrays for Millimeter-Wave Applications : (Invited) 13

学校读者我要写书评

暂无评论

Development of Pillbox-Distributed-Fed Antenna Arrays for Mi...

13th International Conference on Microwave and Millimeter Wave Technology, ICMMT 2021

作者： Wong, Hang Sun, Guang-Hua State Key Lab. Of Terahertz And Millimeter Waves And Guangdong-Hong Kong Jt. Lab. For Big Data Imaging And Commun. Department Of Electrical Engineering City University Of Hong Kong Hong Kong Hong Kong

ISBN: (纸本)9781665434379

This paper introduced the pillbox-distributed feed technique for the wideband millimeter-wave antenna arrays. In one application example, the magneto-electric dipole combined with the patch antenna is adopted as the antenna element, and is excited by the pillbox-distributed feed network. This array achieves an impedance bandwidth up to 30% and a maximum gain of 25 dBi. In another example, the array with side-lobe suppression are demonstrated by integrating the pillbox-distributed network with the non-uniform full-corporate network. Besides, the multi-beam radiation is obtained in this array by placing multiple ports in the focal plane of the parabolic reflector. © 2021 IEEE.

关键词： Antenna feeders

DRAG: Dynamic Region-Aware GCN for Privacy-Leaking Image Detection

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Yang, Guang Cao, Juan Sheng, Qiang Qi, Peng Li, Xirong Li, Jintao Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China Key Lab of Data Engineering and Knowledge Engineering Renmin University of China China

The daily practice of sharing images on social media raises a severe issue about privacy leakage. To address the issue, privacy-leaking image detection is studied recently, with the goal to automatically identify images that may leak privacy. Recent advance on this task benefits from focusing on crucial objects via pretrained object detectors and modeling their correlation. However, these methods have two limitations: 1) they neglect other important elements like scenes, textures, and objects beyond the capacity of pretrained object detectors;2) the correlation among objects is fixed, but a fixed correlation is not appropriate for all the images. To overcome the limitations, we propose the Dynamic Region-Aware Graph Convolutional Network (DRAG) that dynamically finds out crucial regions including objects and other important elements, and models their correlation adaptively for each input image. To find out crucial regions, we cluster spatially-correlated feature channels into several region-aware feature maps. Further, we dynamically model the correlation with the self-attention mechanism and explore the interaction among the regions with a graph convolutional network. The DRAG achieved an accuracy of 87% on the largest dataset for privacy-leaking image detection, which is 10 percentage points higher than the state of the art. The further case study demonstrates that it found out crucial regions containing not only objects but other important elements like textures. The code and more details are in https://***/guang-yanng/DRAG. © 2022, CC BY.

关键词： Object detection

Estimating Noisy Class Posterior with Part-level lab.ls for Noisy lab.l Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhao, Rui Shi, Bin Ruan, Jianfei Pan, Tianze Dong, Bo School of Computer Science and Technology Xi’an Jiaotong University China Shaanxi Province Key Lab of Big Data Knowledge Engineering Xi’an Jiaotong University China School of Physics Xi’an Jiaotong University China School of Distance Education Xi’an Jiaotong University China

In noisy lab.l learning, estimating noisy class posteriors plays a fundamental role for developing consistent classifiers, as it forms the basis for estimating clean class posteriors and the transition matrix. Existing methods typically learn noisy class posteriors by training a classification model with noisy lab.ls. However, when lab.ls are incorrect, these models may be misled to overemphasize the feature parts that do not reflect the instance characteristics, resulting in significant errors in estimating noisy class posteriors. To address this issue, this paper proposes to augment the supervised information with part-level lab.ls, encouraging the model to focus on and integrate richer information from various parts. Specifically, our method first partitions features into distinct parts by cropping instances, yielding part-level lab.ls associated with these various parts. Subsequently, we introduce a novel single-to-multiple transition matrix to model the relationship between the noisy and part-level lab.ls, which incorporates part-level lab.ls into a classifier-consistent framework. Utilizing this framework with part-level lab.ls, we can learn the noisy class posteriors more precisely by guiding the model to integrate information from various parts, ultimately improving the classification performance. Our method is theoretically sound, while experiments show that it is empirically effective in synthetic and real-world noisy benchmarks. Copyright © 2024, The Authors. All rights reserved.

关键词： Contrastive Learning

An Improved YOLOv3 Network for PCB Defect Detection

学校读者我要写书评

暂无评论

An Improved YOLOv3 Network for PCB Defect Detection

2021 China Automation Congress, CAC 2021

作者： Shao, Guifang Zheng, Yuyun Luo, Jincai Gao, Fengqiang Li, Tiejun Ma, Jianqi Wu, Jianyi Liu, Tundong Department of Automation Xiamen Key Lab. of Big Data Intelligent Analysis and Decision Xiamen University Xiamen China School of Aerospace Engineering Xiamen University Xiamen China School of Information Engineering Jimei University Xiamen China

ISBN: (纸本)9781665426473

Different defects with various features and shapes may appeared in one printed circuit board (PCB), making a great challenge for defect detection. To deal with the contradiction between the speed and accuracy of current deep learning networks, we put forward a MobileNet-YOLOv3 network for PCB defect detection. The MobileNet is introduced to replace the backbone network Darknet-53 of YOLOv3, so as to obtain a lightweight network. Meanwhile, the feature pyramid network (FPN) of YOLOv3 is improved by channel reorganization and down-sampling to obtain higher accuracy. Experiments on public data set images with 6 types of defects are conducted, comparative results reveal that the proposed MobileNet-YOLOv3 can achieve higher accuracy than YOLOv3 and YOLOv3-Tiny. According to the frames per second factor compared with YOLOv3-Tiny and YOLOv3, the proposed network is faster than YOLOv3. © 2021 IEEE

关键词： Printed circuit boards

QGEval: Benchmarking Multi-dimensional Evaluation for Question Generation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Fu, Weiping Wei, Bifan Hu, Jianxiang Cai, Zhongmin Liu, Jun School of Computer Science and Technology Xi’an Jiaotong University Xi’an China School of Continuing Education Xi’an Jiaotong University Xi’an China MOE KLINNS Lab School of Automation Science and Engineering Xi’an Jiaotong University Xi’an China Shaanxi Province Key Laboratory of Big Data Knowledge Engineering Xi’an Jiaotong University Xi’an China

Automatically generated questions often suffer from problems such as unclear expression or factual inaccuracies, requiring a reliable and comprehensive evaluation of their quality. Human evaluation is widely used in the field of question generation (QG) and serves as the gold standard for automatic metrics. However, there is a lack of unified human evaluation criteria, which hampers consistent and reliable evaluations of both QG models and automatic metrics. To address this, we propose QGEval, a multi-dimensional Evaluation benchmark for Question Generation, which evaluates both generated questions and existing automatic metrics across 7 dimensions: fluency, clarity, conciseness, relevance, consistency, answerability, and answer consistency. We demonstrate the appropriateness of these dimensions by examining their correlations and distinctions. Through consistent evaluations of QG models and automatic metrics with QGEval, we find that 1) most QG models perform unsatisfactorily in terms of answerability and answer consistency, and 2) existing metrics fail to align well with human judgments when evaluating generated questions across the 7 dimensions. We expect this work to foster the development of both QG technologies and their evaluation. Copyright © 2024, The Authors. All rights reserved.

关键词： Benchmarking