检索结果-内蒙古大学图书馆

International Joint Conference on Neural networks (IJCNN)

作者： Yanfeng Jiang Ning Sun Zhengxian Lu Shuang Peng Yi Zhang Fei Yang Tao Li College of Computer Science Nankai University Tianjin China Tianjin Key Laboratory of Network and Data Security Technology Tianjin China Zhejiang Lab Hangzhou China Haihe Lab of ITAI Tianjin China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Protein language models are currently experiencing a surge in demand owing to their remarkable accuracy in protein structure prediction. Nevertheless, their applications are hindered by the significant computation and memory requirements. The existing optimization strategies primarily focus on computational efficiency while often neglecting memory optimization, thereby restricting their suitability for devices with limited resources. In this paper, we propose MEFold, a novel memory-efficient optimization framework for protein language models that enables efficient inference on resource-constrained devices. MEFold consists of Look-up Table Chunk and Fine-grained Quantization. Look-up Table Chunk reduces the memory of intermediate activations by chunk and avoids the overhead of obtaining the optimal chunk size configuration through pre-computing. For the memory of model parameters, Fine-grained Quantization, delicately controls the scope of quantization to ensure that memory reduction is achieved while preventing declines in accuracy and computational speed. Experimental results show that, compared to the original model, for protein sequences ranging from 74 to 1024 in length, our method significantly reduces the peak memory during inference from 14.7-54.2GB to 6.0-14.4GB, while minimizing the impact on inference latency. On casP14 and CAMEO datasets, the accuracy loss compared to the original model is below 1%. Moreover, our optimization provides various memory-saving alternatives. Our code is available at https://***/llwx593/MEFold.

关键词： Proteins Quantization (signal) Accuracy Computational modeling Memory management Neural networks Predictive models

来源：评论

学校读者我要写书评

暂无评论

DFS-NER: Description Enhanced Few-shot NER via Prompt Learning and Meta-Learning

DFS-NER: Description Enhanced Few-shot NER via Prompt Learni...

引用

2022 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent technology (WI-IAT)

作者： Huinan Huang Yuming Feng Xiaolong Jin Saiping Guan Jiafeng Guo School of Computer Science and Technology University of Chinese Academy of Sciences CAS Key Laboratory of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences

Named Entity Recognition (NER) is a very common task in many social good related domains. Recently, deep learning based NER has gradually matured, but still faces the scarcity problem of labeled data in specific domains. Therefore, researchers focus on few-shot NER to reduce the model’s data dependence and enhance the transferability of the model. However, existing works usually cannot adapt to new entity types and are prone to the so-called negative transfer problem. Therefore, in this paper we propose a type- Description-enhanced Few Shot NER model, called DFS-NER, which effectively integrates the prompt learning paradigm and the meta-learning framework. DFS-NER performs well under frozen pre-training model parameters through continuous templates. It realizes efficient source domain training and target domain parameter fine-tuning through the metalearning framework. We enhance the robustness of entity- type prototype representations by introducing word-word- level and word-type-level contrastive learning objectives and capsule networks as the induction module. Simultaneously, based on discrete prompt learning, a masked-language model learning objective guided by type description is proposed, which can well absorb the semantic information of entity types. Experiments on commonly used datasets, including, SNIPS, Few-NERD, and MIT Movie show that DFS-NER basically surpasses baseline models and achieves the state-of-the-art performance.

关键词： Deep learning Training Adaptation models Analytical models Semantics Prototypes Robustness

来源：评论

学校读者我要写书评

暂无评论

Community discovery method based on complex network of data fusion based on the super network perspective

引用

International Journal of Computers and Applications 2021年第4期43卷 383-390页

作者： Pei, Li School of Computer Science & Technology Xi‘an University of Posts and Telecommunications Xi‘an China Shaanxi Key Laboratory of Network Data Intelligent Processing Xi‘an University of Posts and Telecommunications Xi‘an China Ministry of Education Key Lab For Intelligent Networks and Network Security Xi‘an Jiaotong University Xi‘an China

To enhance the computational efficiency and precision of community discovery, a community discovery algorithm with the mixed label based on the minimum description length (MDI) of information compression is proposed in this paper. Firstly, the community detection is converted to the information compression problem of seeking for an effective network structure and the quality evaluation function is constructed based on the MDL criterion. Secondly, the community discovery algorithm with heuristic mixed label movement is constructed based on the label node movement algorithm and Louvain community addition algorithm so as to reduce the quality evaluation function. Finally, the simulation experiment in the standard test set and API capture Sina microblog dataset shows that the proposed algorithm is superior to the selected comparison algorithm in computational efficiency and precision. © 2018 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

PRADA: Practical Black-Box Adversarial Attacks against Neural Ranking Models

arXiv

引用

arXiv 2022年

作者： Wu, Chen Zhang, Ruqing Guo, Jiafeng de Rijke, Maarten Fan, Yixing Cheng, Xueqi CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China University of Amsterdam Netherlands

Neural ranking models (NRMs) have shown remarkable success in recent years, especially with pre-trained language models. However, deep neural models are notorious for their vulnerability to adversarial examples. Adversarial attacks may become a new type of web spamming technique given our increased reliance on neural information retrieval models. Therefore, it is important to study potential adversarial attacks to identify vulnerabilities of NRMs before they are deployed. In this paper, we introduce the Word Substitution Ranking Attack (WSRA) task against NRMs, which aims to promote a target document in rankings by adding adversarial perturbations to its text. We focus on the decision-based black-box attack setting, where the attackers cannot directly get access to the model information, but can only query the target model to obtain the rank positions of the partial retrieved list. This attack setting is realistic in real-world search engines. We propose a novel Pseudo Relevance-based ADversarial ranking Attack method (PRADA) that learns a surrogate model based on Pseudo Relevance Feedback (PRF) to generate gradients for finding the adversarial perturbations. Experiments on two web search benchmark datasets show that PRADA can outperform existing attack strategies and successfully fool the NRM with small indiscernible perturbations of text. Copyright © 2022, The Authors. All rights reserved.

关键词： Search engines

来源：评论

学校读者我要写书评

暂无评论

AUCSeg: AUC-oriented Pixel-level Long-tail Semantic Segmentation

arXiv

引用

arXiv 2024年

作者： Han, Boyu Xu, Qianqian Yang, Zhiyong Bao, Shilong Wen, Peisong Jiang, Yangbangyan Huang, Qingming 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 Peng Cheng Laboratory China Key Laboratory of Big Data Mining and Knowledge Management CAS China

The Area Under the ROC Curve (AUC) is a well-known metric for evaluating instance-level long-tail learning problems. In the past two decades, many AUC optimization methods have been proposed to improve model performance under long-tail distributions. In this paper, we explore AUC optimization methods in the context of pixel-level long-tail semantic segmentation, a much more complicated scenario. This task introduces two major challenges for AUC optimization techniques. On one hand, AUC optimization in a pixel-level task involves complex coupling across loss terms, with structured inner-image and pairwise inter-image dependencies, complicating theoretical analysis. On the other hand, we find that mini-batch estimation of AUC loss in this case requires a larger batch size, resulting in an unaffordable space complexity. To address these issues, we develop a pixel-level AUC loss function and conduct a dependency-graph-based theoretical analysis of the algorithm’s generalization ability. Additionally, we design a Tail-Classes Memory Bank (T-Memory Bank) to manage the significant memory demand. Finally, comprehensive experiments across various benchmarks confirm the effectiveness of our proposed AUCSeg method. The code is available at https://***/boyuh/AUCSeg. © 2024, CC BY.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

A Discriminative Semantic Ranker for Question Retrieval 21

A Discriminative Semantic Ranker for Question Retrieval

引用

11th ACM SIGIR International Conference on Theory of Information Retrieval, ICTIR 2021

作者： Cai, Yinqiong Fan, Yixing Guo, Jiafeng Zhang, Ruqing Lan, Yanyan Cheng, Xueqi University of Chinese Academy of Sciences Beijing China Cas Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China Institute for Ai Industry Research Tsinghua University Beijing China

ISBN: (纸本)9781450386111

Similar question retrieval is a core task in community-based question answering (CQA) services. To balance the effectiveness and efficiency, the question retrieval system is typically implemented as multi-stage rankers: The first-stage ranker aims to recall potentially relevant questions from a large repository, and the latter stages attempt to re-rank the retrieved results. Most existing works on question retrieval mainly focused on the re-ranking stages, leaving the first-stage ranker to some traditional term-based methods. However, term-based methods often suffer from the vocabulary mismatch problem, especially on short texts, which may block the re-rankers from relevant questions at the very beginning. An alternative is to employ embedding-based methods for the first-stage ranker, which compress texts into dense vectors to enhance the semantic matching. However, these methods often lose the discriminative power as term-based methods, thus introduce noise during retrieval and hurt the recall performance. In this work, we aim to tackle the dilemma of the first-stage ranker, and propose a discriminative semantic ranker, namely DenseTrans, for high-recall retrieval. Specifically, DenseTrans is a densely connected Transformer, which learns semantic embeddings for texts based on Transformer layers. Meanwhile, DenseTrans promotes low-level features through dense connections to keep the discriminative power of the learned representations. DenseTrans is inspired by DenseNet in computer vision (CV), but poses a new way to use the dense connectivity which is totally different from its original design purpose. Experimental results over two question retrieval benchmark datasets show that our model can obtain significant gain on recall against strong term-based methods as well as state-of-the-art embedding-based methods. © 2021 ACM.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Type-Representation Enhanced Event Detection via Structure Encoder

Type-Representation Enhanced Event Detection via Structure E...

引用

Medical Artificial Intelligence (MedAI), IEEE International Conference on

作者： Yuming Feng Linmeng Huang Haozhe Liang Xiaolong Jin Saiping Guan Jiafeng Guo School of Computer Science and Technology University of Chinese Academy of Sciences Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Institute of Systems Engineering Academy of Military Science Beijing

Event detection, aiming to detect whether a sentence contains the type of events of interest, is one of the important tasks in information Extraction. There exist many methods for event detection, which usually treat event detection as a classification problem. These existing methods suffer from the imbalanced training data of different event types. To relieve this problem, this paper proposes a new Type-Representation-Enhanced Event Detection (TREED) method, which introduce hierarchical event type representations to enhance classification accuracy. Specifically, TREED first models the hierarchical relationship among event types in the dataset; Next, upon interactions between the representations of trigger words and event types, it obtains the new representations of trigger words integrating event type information. Finally, TREED classifies the events based on the obtained more accurate trigger word representations. Experimental results on a commonly used benchmark dataset, ACE-2005, demonstrate that the proposed model can improve both accuracy and recall of the event detection task.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Focus on the Stability of Large Systems: Toward Automatic Prediction and Analysis of Vulnerability Threat Intelligence 6

Focus on the Stability of Large Systems: Toward Automatic Pr...

引用

6th IEEE International Conference on data science in Cyberspace, DSC 2021

作者： Song, Shiwen Wu, Qiong Zheng, Xin Wang, Peng Dou, Yuchen Li, Zhongwen Zhai, Lidong Institute Of Information Engineering Chinese Academy Of Sciences BeiJing China Bag Data Academy Zhongke Beijing China Institute Of Computing Technology Chinese Academy Of Sciences Cas Key Laboratory Of Network Data Science And Technology Beijing China

ISBN: (纸本)9781665418157

With the increase in the number of users and business volume, the business systems of Internet companies are becoming more and more complex, resulting in a surge in the number of alarms. A large number of dirty alarms add a huge workload to security operations, which indirectly pose a large number of threats to business systems. At present, most systems use the method of accessing third-party Threat Intelligence to assist operators to realize automatic handling of alarms. However, this method has lagging and accuracy problems, making this work always difficult to meet the requirements of fast and accurate. This article proposes a new method for gathering vulnerability Threat Intelligence, which can obtain vulnerability information in advance of security announcements issued by security vendors. By analyzing the vulnerability disclosure process, this method obtains vulnerability information from the original source submitted by open source mail group, of developers. We used NLP technology and XGBoost model to automatically analyze the vulnerability information, and finally generate FINTEL. The experimental result shows that this method has an accuracy of 93%, and can obtain vulnerability information 10h to 7 days before security vendors release. The scope of application covers all open source code repositories and some closed source repositories. © 2021 IEEE.

关键词： Open source software

来源：评论

学校读者我要写书评

暂无评论

UAV-Assisted Covert data Collection: When Remote Monitoring Meets Covert Communication

UAV-Assisted Covert Data Collection: When Remote Monitoring ...

引用

International Conference on Computer and Communications (ICCC)

作者： Huasheng Su Xiaodong Xu Jing Zhao Jiaxi Zhou Key Lab. of WOC-CAS University of Science and Technology of China Hefei Anhui China Institute of Space Integrated Ground Network Hefei Anhui China

Unmanned aerial vehicle (UAV) assisted remote monitoring and data collection are prevailing applications thanks to its deployment flexibility and strong mobility. To meet security and privacy requirements, this work investigates a UAV-assisted covert data collection scenario, where the collection scheduling, three-dimensional trajectory and transmit power of the involved UAV are jointly optimized while communicating with another UAV receiver in the presence of adversaries. Taking into account the channel uncertainty at the transmitter, the optimization problem is first formulated to maximize the average covert communication rate for the worst case and the average case, respectively. Then, due to the mixed-integer non-convex nature of the optimization problem, the penalty successive convex approximation (P-SCA) scheme is introduced as a solver in association with proper mathematical approximation of some constraints. The numerical results are finally provided to validate the covertness performance of the proposed scheme, which outperforms the baseline who has no information of channel mismatch.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Software Module Clustering Using the Hierarchical Clustering Combination Method 7

Software Module Clustering Using the Hierarchical Clustering...

引用

7th International Conference on Cloud Computing and Big data Analytics, ICCCBDA 2022

作者： Xia, Hong Zhang, Yongkang Chen, Yanping Zhang, Hengshan Wang, Zhongmin Wang, Fengwei School of Computer Science and Technology Xi'An Univ. of P. and Telecom. Shaanxi Key Lab. of Network Data Analysis and Intelligent Processing Xi'an China Xi'An University of Posts and Telecommunications School of Computer Science and Technology Xi'an China ZTECorporation Shenzhen China

ISBN: (纸本)9781665487115

The scale of software applications has increased dramatically. Hierarchical clustering is a good method for modular recovery of software architecture. Because the different evaluation criteria of software types and clustering results, a single hierarchical clustering algorithm cannot integrate the number of clusters, arbitrary decision-making, recovery quality and other indicators on different evaluation standards, there is no comprehensive clustering algorithm selection method. We propose a hierarchical clustering combination algorithm and use principal component analysis to combine the results of multiple hierarchical clusters, the combined result retain the basic information of each clustering algorithm as much as possible. The experimental results show that compared with the single software hierarchical clustering method and other combined methods, the result of our method is closer to expert decomposition, and it has good performance on a variety of indicators. © 2022 IEEE.

关键词： Cloud computing Software architecture Clustering methods Software algorithms Decision making Clustering algorithms Big data

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：