检索结果-内蒙古大学图书馆

A prompt-based approach to adversarial example generation and robustness enhancement

Frontiers of Computer science 2024年第4期18卷 85-96页

作者： Yuting YANG Pei HUANG Juan CAO Jintao LI Yun LIN Feifei MA Key Lab of Intelligent Information Processing of Chinese Academy of Sciences(CAS) Institute of Computing TechnologyCASBeijing 100190China School of Computer Science and Technology University of Chinese Academy of SciencesBeijing 100049China Department of Computer Science Stanford UniversityCA 94305USA School of Computing National University of SingaporeSingapore 119077Singapore Laboratory of Parallel Software and Computational Science Institute of SoftwareChinese Academy of SciencesBeijing 100190China

Recent years have seen the wide application of natural language processing(NLP)models in crucial areas such as finance,medical treatment,and news media,raising concerns about the model robustness and *** find that prompt paradigm can probe special robust defects of pre-trained language *** prompt texts are first constructed for inputs and a pre-trained language model can generate adversarial examples for victim models via *** results show that prompt paradigm can efficiently generate more diverse adversarial examples besides synonym ***,we propose a novel robust training approach based on prompt paradigm which incorporates prompt texts as the alternatives to adversarial examples and enhances robustness under a lightweight minimax-style optimization *** on three real-world tasks and two deep neural models show that our approach can significantly improve the robustness of models to resist adversarial attacks.

关键词： robustness adversarial example prompt learning pre-trained language model

来源：评论

学校读者我要写书评

暂无评论

A new probabilistic algorithm for approximate model counting and extensions for numeric domains 2

A new probabilistic algorithm for approximate model counting...

引用

2nd Workshop on Logics for Reasoning about Preferences, Uncertainty, and Vagueness, PRUV 2018

作者： Ge, Cunjing Ma, Feifei Liu, Tian Zhang, Jian Ma, Xutong State Key Laboratory of Computer Science Institute of Software Chinese Academy of Sciences China Laboratory of Parallel Software and Computational Science Institute of Software Chinese Academy of Sciences China University of Chinese Academy of Sciences China School of Electronics Engineering and Computer Science Peking University China Technology Center of Software Engineering Institute of Software Chinese Academy of Sciences China

Constrained counting is important in domains ranging from artificial intelligence to software analysis. There are already a few approaches for counting models over various types of constraints. Recently, hashing-based approaches achieve success but still rely on solution enumeration. In the IJCAR version, a probabilistic approximate model counter is proposed, which is also a hashing-based universal framework, but with only satisfiability queries. A dynamic stopping criteria, for the algorithm, is presented, which has not been studied yet in previous works of hashing-based approaches. Although the algorithm lacks theoretical guarantee, it works well in practice. In this paper, we further extend our approach to SMT(LIA) formulas with a new encoding technique. © 2018 CEUR-WS. All Rights Reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Sleep scheduling method based on half-sleep state in the distributed sensor network 9

Sleep scheduling method based on half-sleep state in the dis...

引用

9th International ICST Conference on Testbeds and Research Infrastructures for the Development of Networks and Communities, TridentCom 2014

作者： Deng, Pan Chen, Feng Yan, Biying Zhang, Jianwei Zhao, Long Wan, Jiafu Lab. of Parallel Software and Computational Science Institute of Software Chinese Academy of Sciences Beijing China State Key Laboratory of Software Development Environment Beihang University Beijing China College of Electrical Engineering and Automation Jiangxi University of Science and Technology Ganzhou China

ISBN: (纸本)9783319133256

In order to extend the sensor’s lifetime, this paper researched deeply into the sleep scheduling mechanism in the distributed sensor network. Now, the commonly used sleep scheduling methods based on the coverage have the problem of response delay, so we proposed a sleep scheduling method based on the half-sleep state to overcome this shortcoming. Under the control of regional agent node, this method adopts a minimum coverage algorithm based on the approximate solution to select the minimum coverage node set. Experimental results show that the proposed scheduling method can both reduce power consumption of the whole network effectively and extend the lifetime of the sensor noticeably. © Institute for Computer sciences, Social Informatics and Telecommunications Engineering 2014.

关键词： Sensor networks

来源：评论

学校读者我要写书评

暂无评论

A BPEL-based web service flow engine in the early warning of the volcano effusion 4th

A BPEL-based web service flow engine in the early warning of...

引用

4th International Conference on Cloud Computing, CloudComp 2013

作者： Pang, Jingyuan Wang, Chen Deng, Pan Lu, Yanhong Liu, Hao Earthquake Administration of Jilin Province Changchun30117 China Institute of Geophysics China Earthquake Administration Beijing China Laboratory of Parallel Software and Computational Science Institute of Software Chinese Academy of Sciences Beijing100190 China

ISBN: (纸本)9783319055053

Web service flow engine is the core of the service assembling and cooperating system in volcano early warning data-shared platform. We design and realize a lightweight web service flow based on BPEL standard. It takes charge of parsing the service flow model constructed by users, checking-up the syntax and logic of the service flow file, constructing the basic elements of the given service flow, and then executing it according to the BPEL standard. Then the engine collects all the executing information and invoking results, at the same time it also supervises the execution of web services, managing the invocation of services. Finally it returns the results back to users. Experimental results show that our engine can work correctly, the time efficiency is high and the occupancy rate of system resources is nearly zero. © Institute for Computer sciences, Social Informatics and Telecommunications Engineering 2014.

关键词： Web services

来源：评论

学校读者我要写书评

暂无评论

Online Bayesian max-margin subspace multi-view learning 25

Online Bayesian max-margin subspace multi-view learning

引用

25th International Joint Conference on Artificial Intelligence, IJCAI 2016

作者： He, Jia Du, Changying Zhuang, Fuzhen Yin, Xin He, Qing Long, Guoping Institute of Computing Technology CAS Beijing100190 China Laboratory of Parallel Software and Computational Science Institute of Software Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China

Last decades have witnessed a number of studies devoted to multi-view learning algorithms, however, few efforts have been made to handle online multi-view learning scenarios. In this paper, we propose an online Bayesian multi-view learning algorithm to learn predictive subspace with max-margin principle. Specifically, we first define the latent margin loss for classification in the subspace, and then cast the learning problem into a variational Bayesian framework by exploiting the pseudo-likelihood and data augmentation idea. With the variational approximate posterior inferred from the past samples, we can naturally combine historical knowledge with new arrival data, in a Bayesian Passive-Aggressive style. Experiments on various classification tasks show that our model have superior performance.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Multi-mode network analysis under differential privacy

Multi-mode network analysis under differential privacy

引用

2021 Guangdong-Hong Kong-Macao Greater Bay Area Artificial Intelligence and Big Data Forum, AIBDF 2021

作者： Song, Yuning Ding, Liping Dong, Mengying Liu, Xuehua Wang, Xiao Laboratory of Parallel Software and Computational Science Institute of Software Chinese Academy of Sciences Beijing100190 China Guangzhou511458 China State Key Laboratory of Information Security Institute of Information Engineering Chinese Academy of Sciences Beijing100093 China

With the advent of the big data era and the advancement of social network analysis, the public is increasingly concerned about the privacy protection in today's complex social networks. For the past few years, the rapid development of differential privacy (DP) technology, as a method with a reliable theoretical basis, can effectively solve the key problem of how to "disassociate" personal information in social networks. This paper focuses on the multi-mode heterogeneous network model which has attracted a lot of attention in the field of network research. It introduces differential privacy and its application in big social networks briefly first, and then proposes a centrality-analysis method based on DP in a typical social network, i.e. the multi-mode network. The calculation principle and applicable scenarios are discussed. Then, its utility is analyzed and evaluated through experimental simulation. Possible improvement of DP algorithm in multi-mode networks above is prospected in the end. © 2021 Institute of Physics Publishing. All rights reserved.

关键词： Heterogeneous networks

来源：评论

学校读者我要写书评

暂无评论

Learning beyond predefined label space via bayesian nonparametric topic modelling 1

引用

15th European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, ECML PKDD 2016

作者： Du, Changying Zhuang, Fuzhen He, Jia He, Qing Long, Guoping Institute of Computing Technology CAS Beijing100190 China Laboratory of Parallel Software and Computational Science Institute of Software Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China

ISBN: (数字)9783319461281

ISBN: (纸本)9783319461274

In real world machine learning applications, testing data may contain some meaningful new categories that have not been seen in labeled training data. To simultaneously recognize new data categories and assign most appropriate category labels to the data actually from known categories, existing models assume the number of unknown new categories is pre-specified, though it is difficult to determine in advance. In this paper, we propose a Bayesian nonparametric topic model to automatically infer this number, based on the hierarchical Dirichlet process and the notion of latent Dirichlet allocation. Exact inference in our model is intractable, so we provide an efficient collapsed Gibbs sampling algorithm for approximate posterior inference. Extensive experiments on various text data sets show that: (a) compared with parametric approaches that use pre-specified true number of new categories, the proposed nonparametric approach can yield comparable performance;and (b) when the exact number of new categories is unavailable, i.e. the parametric approaches only have a rough idea about the new categories, our approach has evident performance advantages. © Springer International Publishing AG 2016.

关键词： Inference engines

来源：评论

学校读者我要写书评

暂无评论

CLSIFT: An optimization study of the scale invariance feature transform on GPUs

CLSIFT: An optimization study of the scale invariance featur...

引用

15th IEEE International Conference on High Performance Computing and Communications, HPCC 2013 and 11th IEEE/IFIP International Conference on Embedded and Ubiquitous Computing, EUC 2013

作者： Wang, Weiyan Zhang, Yunquan Guoping, Long Yan, Shengen Jia, Haipeng Lab. of Parallel Software and Computational Science Institute of Software Chinese Academy of Sciences China State Key Laboratory of Computing Science Institute of Software Chinese Academy of Sciences China School of Computer and Control Engineering University of Chinese Academy of Sciences Beijing China

ISBN: (纸本)9780769550886

Scale Invariance Feature Transform (SIFT) is quite suitable for image matching because of its invariance to image scaling, rotation and slight changes in illumination or viewpoint. However, due to high computation complexity it's technically challenging to deploy SIFT in real time application situations. To address this problem, we propose CLSIFT, an OpenCL based highly speeded up and performance portable SIFT solution. Important optimization techniques employed in CLSIFT such as: (1) For less global memory traffic, independent logical functions are merged into the same kernel to reuse data.(2) loop buffers are introduced in for data and intermediate results reusing.(3)Task queue used to schedule threads in the same branch to remove branch divergences. (4) Data partition is based on the statics patterns for workload balance among workgroups. (5) Overlap of CPU time and better parallel strategies are used too. With all mentioned efforts, CLSIFT processes lena. jpg at 74.2 FPS and 43.4FPS respectively on NVidia and AMD GPUS, much higher than CPU's nearly 10 FPS and the known fastest SIFTGPU's 39.8 FPS and 13FPS. Moreover in a quantitative comparison approach we analyze those successful strategies beating SIFTGPU, a famous existing GPU implementation. Additionally, we observe and conclude that NVidia GPU achieves better occupancy and performance due to some factors. Finally, we summarize some techniques and empirical guiding principles that may be shared by other applications on GPU. © 2013 IEEE.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

Modeling the locality in graph traversals

Modeling the locality in graph traversals

引用

41st International Conference on parallel Processing, ICPP 2012

作者： Yuan, Liang Ding, Chen Štefankovič, Daniel Zhang, Yunquan Lab. of Parallel Software and Computational Science Institute of Software Chinese Academy of Sciences China State Key Laboratory of Computer Science Institute of Software Chinese Academy of Sciences China Graduate University of Chinese Academy of Sciences China Computer Science Department University of Rochester United States

ISBN: (纸本)9780769547961

An increasing number of applications in physical and social sciences require the analysis of large graphs. The efficiency of these programs strongly depends on their memory usage especially the locality of graph data access. Intuitively, the locality in computation should reflect the locality in graph topology. Existing locality models, however, operate either at program level for regular loops and arrays or at trace level for arbitrary access streams. They are not sufficient to characterize the relation between locality and connectivity. This paper presents a new metrics called the vertex distance and uses it to model the locality in breadth-first graph traversal (BFS). It shows three models that use the average node degree and the edge distribution to predict the number of BFS levels and the reuse distance distribution of BFS. Finally, it evaluates the new models using random and non-random graphs. © 2012 IEEE.

关键词： Graph theory

来源：评论

学校读者我要写书评

暂无评论

A capability-based matchmaking mechanism supporting resource aggregation within large-scale distributed computing infrastructures 4th

A capability-based matchmaking mechanism supporting resource...

引用

4th International Conference on Cloud Computing, CloudComp 2013

作者： Liang, Feng Liu, Hai Liu, Yunzhen Ma, Shilong Zheng, Siyao Deng, Pan State Key Lab of Software Development Environment Beihang University Beijing100191 China The Laboratory of Embedded Systems Beihang University Beijing100191 China Lab of Parallel Software and Computational Science Institute of Software Chinese Academy of Sciences Beijing100190 China

ISBN: (纸本)9783319055053

Facing the large-scale, heterogeneous dynamic resource and the complex constraints of computation-intensive parallel scientific applications, collaborating large-scale computation resource for these applications within Large-Scale Distributed Computing Infrastructures like grids and clouds are challenging. This paper addresses this issue by proposing a Resource Capability Model and implementing a corresponding language GSO and an information service Application Information Service to ensure not only the resource description, aggregation, matchmaking but also life cycle management of the reservations and tasks. Experiment result shows that our mechanism is scalable in both Grid and Cloud environment. © Institute for Computer sciences, Social Informatics and Telecommunications Engineering 2014.

关键词： Information services

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：