检索结果-内蒙古大学图书馆

48th IEEE International Conference on Acoustics, Speech and Signal processing, ICASSP 2023

作者： Zhang, Lihua Liu, Quan Huang, Zhigang Wu, Lan Soochow University School of Computer Science and Technology Suzhou China Soochow University Provincial Key Laboratory for Computer Information Processing Technology Suzhou China

ISBN: (纸本)9781728163277

A powerful method for automated decision systems is Adversarial Imitation Learning (AIL). It is based on a generative adversarial framework that alternately optimizes a generator (learner) and a discriminator (reward function). In the popular mind, a high-accuracy discriminator results in informative rewards, thus AIL must balance the performance between a generator and a discriminator. However, we find that the AIL reward function is biased and also results in informative rewards. Thus, we theoretically analyze the bias in the AIL reward function and find that balancing the performance of a generator and a discriminator is not necessary when we recover an unbiased reward function. Further, we propose a mutual information based auxiliary reward function. Experiments on continuous control tasks indicate that MI-GAIL is able to address the bias problem of the AIL reward function and further improve sample efficiency and training stability compared with up-to-date algorithms. © 2023 IEEE.

关键词： Adversarial imitation learning Continuous control Mutual information Potential-based reward function Reward bias

来源：评论

学校读者我要写书评

暂无评论

Reliable neighbors-based collaborative filtering for recommendation systems 1

引用

1st International Conference on Neural Computing for Advanced Applications, NCAA 2020

作者： Zhang, Li Sun, Xiaohan Cheng, Jiangfei Li, Zepeng School of Computer Science and Technology Joint International Research Laboratory of Machine Learning and Neuromorphic Computing Soochow University SuzhouJiangsu215006 China Provincial Key Laboratory for Computer Information Processing Technology Soochow University SuzhouJiangsu215006 China

ISBN: (数字)9789811576706

ISBN: (纸本)9789811576690

The sparsity of data and the diversity of ratings have a great effect on the performance of recommendation systems. To deal with these two issues, this paper proposes a reliable neighbors-based collaborative filtering (RNCF) method for recommendation systems. RNCF first uses the item similarity to pre-estimated missing ratings to solve the issue of data sparsity, which enriches the sample data and reduces the impact of extremely sparse ratings on the recommendation effect. Then, RNCF finds a set of reliable users for each item by defining a reliable degree for each item. Next, RNCF adopts the K-means clustering algorithm to find the similar users for a target user. The set of reliable neighbors is the intersection of sets of reliable users and similar users. Finally, RNCF provides a predicted rating of a target user on some item using the pre-estimated ratings of corresponding reliable neighbors. Experimental analysis shows that RNCF can improve the performance at the extreme levels of sparsity and rating diversity in recommendation systems. © Springer Nature Singapore Pte Ltd. 2020.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Exploiting constituent dependencies for tree kernel-based semantic relation extraction

Exploiting constituent dependencies for tree kernel-based se...

引用

22nd International Conference on Computational Linguistics, Coling 2008

作者： Qian, Longhua Zhou, Guodong Kong, Fang Zhu, Qiaoming Qian, Peide Jiangsu Provincial Key Lab. for Computer Information Processing Technology School of Computer Science and Technology Soochow University 1 Shizi Street Suzhou 215006 China

ISBN: (纸本)9781905593446

This paper proposes a new approach to dynamically determine the tree span for tree kernel-based semantic relation extraction. It exploits constituent dependencies to keep the nodes and their head children along the path connecting the two entities, while removing the noisy information from the syntactic parse tree, eventually leading to a dynamic syntactic parse tree. This paper also explores entity features and their combined features in a unified parse and semantic tree, which integrates both structured syntactic parse information and entity-related semantic information. Evaluation on the ACE RDC 2004 corpus shows that our dynamic syntactic parse tree outperforms all previous tree spans, and the composite kernel combining this tree kernel with a linear state-of-the-art feature-based kernel, achieves the so far best performance. © 2008. Licensed under the Creative Commons.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Robust Monocular Visual Odometry using Optical Flows for Mobile Robots 35

Robust Monocular Visual Odometry using Optical Flows for Mob...

引用

第35届中国控制会议

作者： LI Haifeng HU Zunhe CHEN Xinwei College of Computer Science and Technology Civil Aviation University of China Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Department of Computer Science Minjiang University

ISBN: (纸本)9781509009107

A robust optical flow-based visual odometry method using a single onboard camera is proposed in this *** improve the quality of the noisy optical flows,a correction method across multiple frames is ***,the optical flows in the plane at infinity are detected and removed as these optical flows have very low signal to noise ratio for robot translation ***,a RANSAC approach for robot ego-motion estimation is *** experiments are carried out and the results show that the proposed method is able to estimate the camera trajectory robustly with reasonable accuracy.

关键词： visual odometry optical flows mobile robot

来源：评论

学校读者我要写书评

暂无评论

Research of event pronoun resolution

Research of event pronoun resolution

引用

2011 International Conference on Asian Language processing, IALP 2011

作者： Zhang, Ning Kong, Fang Li, Peifeng JiangSu Provincial Key Lab. for Computer Information Processing Technology Soochow University Suzhou China School of Computer Science and Technology Soochow University Suzhou China

ISBN: (纸本)9780769545547

Event anaphora resolution plays an important role in discourse analysis. In comparison with general noun phrases, pronouns carry little information of themselves, resolving the event pronouns is a more difficult task. This paper proposes a machine learning-based framework for event pronoun resolution. All kinds of features, including both flat and structural features, are explored for event pronoun resolution. Experiments on OntoNotes corpus show that both flat and structural features are very effective for this task. © 2011 IEEE.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Semi-supervised feature selection using sparse laplacian support vector machine 1

引用

1st International Conference on Neural Computing for Advanced Applications, NCAA 2020

作者： Zhang, Li Zheng, Xiaohan Xu, Zhiqiang School of Computer Science and Technology Joint International Research Laboratory of Machine Learning and Neuromorphic Computing Soochow University SuzhouJiangsu215006 China Provincial Key Laboratory for Computer Information Processing Technology Soochow University SuzhouJiangsu215006 China

ISBN: (数字)9789811576706

ISBN: (纸本)9789811576690

Semi-supervised feature selection is an active topic in machine learning and data mining. Laplacian support vector machine (LapSVM) has been successfully applied to semi-supervised learning. However, LapSVM cannot be directly applied to feature selection. To remedy it, we propose a sparse Laplacian support vector machine (SLapSVM) and apply it to semi-supervised feature selection. On the basis of LapSVM, SLapSVM introduces the 1-norm regularization, which means the solution of SLapSVM has sparsity. In addition, the training procedure of SLapSVM can be formulated as solving a quadratic programming problem, which indicates that the solution of SLapSVM is unique and global. SLapSVM can perform feature selection and classification at the same time. Experimental results on semi-supervised classification problems show the feasibility and effectiveness of the proposed semi-supervised learning algorithms. © Springer Nature Singapore Pte Ltd. 2020.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

An improved Isomap method for manifold learning

引用

International Journal of Intelligent Computing and Cybernetics 2017年第1期10卷 30-40页

作者： Taiguo Qu Zixing Cai College of Computer Science and Technology Hengyang Normal UniversityHengyangChina Hunan Provincial Key Laboratory of Intelligent Information Processing and Application HengyangChina School of Information Science and Engineering Central South UniversityChangshaChina

Purpose-Isometric feature mapping(Isomap)is a very popular manifold learning method and is widely used in dimensionality reduction and data *** most time-consuming step in Isomap is to compute the shortest paths between all pairs of data points based on a neighbourhood *** classical Isomap(C-Isomap)is very slow,due to the use of Floyd’s algorithm to compute the shortest *** purpose of this paper is to speed up ***/methodology/approach-Through theoretical analysis,it is found that the neighbourhood graph in Isomap is *** this case,the Dijkstra’s algorithm with Fibonacci heap(Fib-Dij)is faster than Floyd’s *** this paper,an improved Isomap method based on Fib-Dij is *** using Fib-Dij to replace Floyd’s algorithm,an improved Isomap method is presented in this ***-Using the S-curve,the Swiss-roll,the Frey face database,the mixed national institute of standards and technology database of handwritten digits and a face image database,the performance of the proposed method is compared with C-Isomap,showing the consistency with C-Isomap and marked improvements in terms of the high *** also demonstrate that Fib-Dij reduces the computation time of the shortest paths from O(N3)to O(N2lgN).Research limitations/implications-Due to the limitations of the computer,the sizes of the data sets in this paper are all smaller than 3,***,researchers are encouraged to test the proposed algorithm on larger data ***/value-The new method based on Fib-Dij can greatly improve the speed of Isomap.

关键词： Dijkstra’s algorithm Fibonacci heap Floyd’s algorithm Isomap Manifold learning Shortest path Paper type Research paper

来源：评论

学校读者我要写书评

暂无评论

Research on cross-document coreference of Chinese person name

Research on cross-document coreference of Chinese person nam...

引用

2011 International Conference on Asian Language processing, IALP 2011

作者： Ni, Ji Kong, Fang Li, Peifeng Zhu, Qiaoming JiangSu Provincial Key Lab. for Computer Information Processing Technology Soochow University Suzhou China School of Computer Science and Technology Soochow University Suzhou China

ISBN: (纸本)9780769545547

In reality, different persons often have the same person name. The Person Cross Document Co-reference Resolution is a task, which requires that all and only the textual mentions of an entity of type Person be individuated in a collection of text documents. In this paper, we implement a Chinese Person Name Cross Document Co-reference Resolution System. First, we utilize name identification module to recognize all person names of the texts, and then classify the document collection of same person name by rules preliminarily, and at last, we compute similarities of each classification based on VSM, according to the prior similarities, the system get the final classification results. We test the system on 30 usual Chinese names of the corpus provided by CLP, and average F measure is 85.9%. © 2011 IEEE.

关键词： information retrieval systems

来源：评论

学校读者我要写书评

暂无评论

Answering contextual questions based on ontologies and question templates

引用

Frontiers of computer Science 2011年第4期5卷 405-418页

作者： Dongsheng WANG Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China Graduate University of the Chinese Academy of Sciences Beijing 100049 China Department of Computer Science Jiangsu University of Science and Technology Zhenjiang 212003 China

Contextual question answering （CQA）, in which user information needs are satisfied through an interactive question answering （QA） dialog, has recently attracted more research attention. One challenge is to fuse contextual information into the understanding process of relevant questions. In this paper, a discourse structure is proposed to maintain semantic information, and approaches for recognition of relevancy type and fusion of contextual information according to relevancy type are proposed. The system is evaluated on real contextual QA data. The results show that better performance is achieved than a baseline system and almost the same performance as when these contextual phenomena are resolved manually. A detailed evaluation analysis is presented.

关键词： Contextual question answering (CQA),ontology question templates

来源：评论

学校读者我要写书评

暂无评论

Research of noun phrase coreference resolution

Research of noun phrase coreference resolution

引用

2011 International Conference on Asian Language processing, IALP 2011

作者： Gao, Junwei Kong, Fang Li, Peifeng Zhu, Qiaoming JiangSu Provincial Key Lab. for Computer Information Processing Technology Soochow University Suzhou China School of Computer Science and Technology Soochow University Suzhou China

ISBN: (纸本)9780769545547

Coreference resolution is an important subtask in natural language processing systems. The process of it is to find whether two expressions in natural language refer to the same entity in the world. Machine learning approaches to this problem have been reasonably successful, operating primarily by recasting the problem as a classification task. A great deal of research has been done on this task in English, using approaches ranging from those based on linguistics to those based on machine learning. In Chinese, however, much less work has been done in this area. The lack of public resources is a big problem in the research of Chinese NLP. The other problem is that some features are more difficult to get than those features of English. In this paper, We present a noun phrase coreference system that refers to the work of Soon et al. (2001). We also explore the impact of various features on our system's performance. Experiments on the Chinese portion of OntoNotes 3.0 show that the platform achieves a good performance. © 2011 IEEE.

关键词： Natural language processing systems

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：