检索结果-内蒙古大学图书馆

Annual Conference on Neural Information Processing systems

作者： Kun Gai Guangyun Chen Changshui Zhang State Key Laboratory on Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology (TNList) Department of Automation Tsinghua University Beijing 100084 China

ISBN: (纸本)9781617823800

In this paper, we point out that there exist scaling and initialization problems in most existing multiple kernel learning (MKL) approaches, which employ the large margin principle to jointly learn both a kernel and an SVM classifier. The reason is that the margin itself can not well describe how good a kernel is due to the negligence of the scaling. We use the ratio between the margin and the radius of the minimum enclosing ball to measure the goodness of a kernel, and present a new minimization formulation for kernel learning. This formulation is invariant to scalings of learned kernels, and when learning linear combination of basis kernels it is also invariant to scalings of basis kernels and to the types (e.g., L_1 or L_2) of norm constraints on combination coefficients. We establish the differentiability of our formulation, and propose a gradient projection algorithm for kernel learning. Experiments show that our method significantly outperforms both SVM with the uniform combination of basis kernels and other state-of-art MKL approaches.

关键词：

来源：评论

学校读者我要写书评

暂无评论

COOPERATION BASED SPECTRUM SHARING IN COGNITIVE RADIO NETWORKS

COOPERATION BASED SPECTRUM SHARING IN COGNITIVE RADIO NETWOR...

引用

2010 IEEE Youth Conference on Information,Computing and Telecommunications(2010 IEEE青年信息、计算和通信技术大会)

作者： Zhong Chen Xianda Zhang TianXiang Luan State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology Department of Automation Tsinghua University Beijing 100-084 China

In this paper, we propose a cooperative spectrum sharing scheme based on decode and forward relaying and a three phase protocol in cognitive radio networks for a secondary system to access the spectrum concurrent with the primary system. The whole system consists of a pair of primary transmitter (PT) and receiver (PR) and a pair of secondary transmitter (ST) and receiver (SR). PT broadcasts the signal to ST and PR during the .rst fraction 1 - α of time. ST decodes and regenerates the relaying signal to PR in the second fraction αβ of time and transmits the secondary signal to SR in the third fraction α(1-β) of time. We consider the case that the channel state information (CSI) is fully known where the secondary system won't impact the primary transmission by ensuring the achievable rate, and the case that the distance information between each nodes is known where outage performance of the primary system is ensured. The performance of the secondary system is optimized to get optimal α and β Simulation results illustrate that the proposed scheme can *** realize spectrum sharing.

关键词： Cognitive radio spectrum sharing decode and forward relaying rate outage probability

来源：评论

学校读者我要写书评

暂无评论

AN OPTIMAL COOPERATIVE SPECTRUM SENSING SCHEME BASED ON LINEAR COHERENT MAC UNDER INTERFERENCE TEMPERATURE CONSTRAINTS

AN OPTIMAL COOPERATIVE SPECTRUM SENSING SCHEME BASED ON LINE...

引用

2010 IEEE Youth Conference on Information,Computing and Telecommunications(2010 IEEE青年信息、计算和通信技术大会)

作者： Zhong Chen Xianda Zhang Liang Zeng State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology Department of Automation Tsinghua University Beijing 100-084 China

In this paper, we consider a cognitive radio (CR) system in non-ideal fading wireless channels and propose a cooperative spectrum sensing scheme based on coherent multiple access channel (MAC) serving as an alternative way to improve the cooperative spectrum sensing performance. We assume that the gains of the observations and transmission channels are all known which is available since we use standard preambleaided channel estimation techniques to require channel state information. The key feature of the proposed scheme is that energy of the observations is transmitted by amplify and forward transmission from cognitive users (CUs) to a fusion center (FC) and a linear combiner at FC. Optimal weighting coef- .cients of the proposed scheme are got by maximizing the detection probability under a target false alarm probability and transmit power constraint. The effectiveness of the proposed scheme is *** by the simulations.

关键词： Cognitive radio cooperative spectrum sensing coherent MAC de.ection coef.cient

来源：评论

学校读者我要写书评

暂无评论

Hybrid forecasting model research on stock data mining

Hybrid forecasting model research on stock data mining

引用

International Conference on New Trends in Information and Service Science, NISS

作者： Fangwen Zhai Qinghua Wen Zehong Yang Yixu Song Department of Computer Science and Technology Tsinghua University Beijing China State Key Laboratory on Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology Beijing China

ISBN: (纸本)9781424469826;9788988678183

The synergy effect's benefit is widely accepted. The object of this paper is to investigate whether a hybrid approach combining different stock prediction approaches together can dramatically outperform the single approach and compare the performance of different hybrid approaches. The hybrid model includes three well-researched algorithms: back propagation neural network (BPNN), adaptive network-based fuzzy neural inference system (ANFIS) and support vector machine (SVM). First, we utilize them independently to single-step forecast the stock price, and then integrate the three forecasts into a final result by a combining strategy. Two different combining methods are investigated. The first method is a linear combination of the three forecasts. The second method combines them by a neural network. We have all of the algorithms experiment on the S&P500 Index. The experiment verifies that by combining the single algorithm appropriately, better performance can be achieved.

关键词： Predictive models Data mining Neural networks Fuzzy systems Artificial neural networks Fuzzy neural networks Adaptive systems Support vector machines Recurrent neural networks Exchange rates

来源：评论

学校读者我要写书评

暂无评论

Homotopy Regularization for Boosting

Homotopy Regularization for Boosting

引用

IEEE International Conference on Data Mining (ICDM)

作者： Zheng Wang Yangqiu Song Changshui Zhang State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory of Information Science and Technology (TNList) Department of Automation Tsinghua University China IBM Research China

In this paper, we present a homotopy regularization algorithm for boosting. We introduce a regularization term with adaptive weight into the boosting framework and compose a homotopy objective function. Optimization of this objective approximately composes a solution path for the regularized boosting. Following this path, we can find suitable solution efficiently using early stopping. Experiments show that this adaptive regularization method gives a more efficient parameter selection strategy than regularized boosting and semi supervised boosting algorithms, and significantly improves the performances of traditional AdaBoost and related methods.

关键词： Boosting Training Algorithm design and analysis Laplace equations Manifolds Tuning Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Tensor rank of the tripartite state |W〉⊗n

引用

Physical Review A 2010年第1期81卷 014301-014301页

作者： Nengkun Yu Eric Chitambar Cheng Guo Runyao Duan State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology Department of Computer Science and Technology Tsinghua University Beijing 100084 China and Centre for Quantum Computation and Intelligent Systems (QCIS) Faculty of Engineering and Information Technology University of Technology Sydney New South Wales 2007 Australia Physics Department University of Michigan 450 Church Street Ann Arbor Michigan 48109-1040 USA

Tensor rank refers to the number of product states needed to express a given multipartite quantum state. Its nonadditivity as an entanglement measure has recently been observed. In this Brief Report, we estimate the tensor rank of multiple copies of the tripartite state |W〉=13(|100〉+|010〉+|001〉). Both an upper bound and a lower bound of this rank are derived. In particular, it is proven that the rank of |W〉⊗2 is 7, thus resolving a previously open problem. Some implications of this result are discussed in terms of transformation rates between |W〉⊗n and multiple copies of the state |GHZ〉=12(|000〉+|111〉).

关键词： Tensors

来源：评论

学校读者我要写书评

暂无评论

An effective trajectory optimization method for robotic belt grinding based on intelligent algorithm

An effective trajectory optimization method for robotic belt...

引用

IEEE International Conference on Robotics and Biomimetics

作者： Wei Liang Yixu Song Hongbo Lv Peifa Jia Zhongxue Gan Lizhe Qi State Key Laboratory on Intelligent Technology and Systems Tsinghua National Laboratory of Information Science and Technology Department of Computer Science and Technology Tsinghua University Beijing China Intersmart Robotic Systems Company Limited Langfang Hebei China

ISBN: (纸本)9781424493197

In this paper, a novel method for robotic belt grinding based on support vector machine and particle swarm optimization algorithm is presented. Firstly, the dynamic model of the robotic belt grinding process is built using support vector machine method. This is the basis of our work because the dynamic model shows the relation between the removal and control parameters (contact force and robot's speed) of robot. Secondly, the method of reverse solution of the dynamic model is introduced. According to this method, control parameters of robot can be accurately calculated by the given value of removal. Thirdly, the standard PSO algorithm is introduced to get smooth and stable trajectories of the control parameters, because the trajectory jitter of the control parameters has a great influence on the grinding accuracy. Finally, a variation on the traditional PSO algorithm is presented, which is called the cooperative particle swarm optimizer, or CPSO, employing cooperative behavior to significantly improve the performance of the original algorithm. The experiment results show that the novel method for robotic belt grinding performs well in the control of the robot parameters and the grinding accuracy and efficiency is improved.

关键词： Trajectory Optimization Support vector machines Belts Mobile robots Mathematical model

来源：评论

学校读者我要写书评

暂无评论

Evolutionary Self-learning Scheduling Approach for Wireless Sensor Network

Evolutionary Self-learning Scheduling Approach for Wireless ...

引用

2010 International Conference on intelligent Computation technology and Automation(2010 智能计算技术与自动化国际会议 ICICTA 2010)

作者： Jianjun Niu State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology Department of Computer Science and Technology Tsinghua University Beijing China Beijing Special Engineering Design and Resear

Energy efficiency is very important for wireless sensor network (WSN). This paper presents an evolutionary self-learning scheduling approach (ESSA) to reduce energy consumption for WSN. The ESSA is based on a new proposed scheme - evolutionary Q-learning with continuousaction (EQC) approach, which combines an extension of Q-learning method with particle swarm optimization (PSO) algorithm. The action space of EQC is partitioned into lots of subintervals. And each endpoint of the subintervals is characterized by a discrete action value and a Q-value. The continuous action value is the weighted average of discrete actions according to their Q values. The PSO algorithm is combined to let an agent profit the experience of other agents. We valid the ESSA in a MAC protocol and simulation results show that the ESSA is an effective method and performs much better than SMAC protocol.

关键词： Wirless Sensor Network Scheduling Q-Learning Particle Swarm Optimization

来源：评论

学校读者我要写书评

暂无评论

Multi-objective dynamic programming algorithm for aircraft arrival sequencing and runway scheduling

引用

Kongzhi Lilun Yu Yingyong/Control Theory and Applications 2010年第7期27卷 827-835页

作者： Ying, Sheng-Gang Sun, Fu-Chun Hu, Lai-Hong Liu, Hua-Ping Zhang, Xue-Jun State Key Laboratory of Intelligent Technology and Systems Tsinghua University Beijing 100084 China The Second Artillery Engineering College Xi'an Shaanxi 710025 China School of Instrumentation Science and Optoelectronics Engineering Beihang University Beijing 100191 China

Optimization of runway scheduling for aircraft landings plays an important role in modern air traffic control, by maximizing throughput of an airport and minimizing fuel cost of aircrafts. As a nondeterministic polynomialcomplete(NP-C) problem, the runway scheduling of a considerable number of aircrafts in a multirunway airport hasn't been effectively solved. Because of considerable computation required by the traditional dynamic programming algorithm under constrained position shifting(CPS), we can only sequence aircrafts and schedule the time of arrival in a single-runway airport. This paper presents a new dynamic programming algorithm by changing the way of recurrence and combining the traditional one with several other methods including a greedy algorithm. Our algorithm can solve the problem of multirunway scheduling with multi-object efficiently and effectively. A large number of experiments show that the complexity of the algorithm is almost linearly proportional to the number of aircrafts, and the algorithm can optimize both throughput and landing cost simultaneously in a short period of time.

关键词： Scheduling

来源：评论

学校读者我要写书评

暂无评论

Locally indistinguishable subspaces spanned by three-qubit unextendible product bases

引用

Physical Review A 2010年第3期81卷 032329-032329页

作者： Runyao Duan Yu Xin Mingsheng Ying Centre for Quantum Computation and Intelligent Systems (QCIS) Faculty of Engineering and Information Technology University of Technology Sydney New South Wales 2007 Australia State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology Department of Computer Science and Technology Tsinghua University Beijing 100084 People's Republic of China Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology Cambridge Massachusetts 02139 USA Department of Physics Tsinghua University Beijing 100084 People's Republic of China

We study the local distinguishability of general multiqubit states and show that local projective measurements and classical communication are as powerful as the most general local measurements and classical communication. Remarkably, this indicates that the local distinguishability of multiqubit states can be decided efficiently. Another useful consequence is that a set of orthogonal n-qubit states is locally distinguishable only if the summation of their orthogonal Schmidt numbers is less than the total dimension 2n. Employing these results, we show that any orthonormal basis of a subspace spanned by arbitrary three-qubit orthogonal unextendible product bases (UPB) cannot be exactly distinguishable by local operations and classical communication. This not only reveals another intrinsic property of three-qubit orthogonal UPB but also provides a class of locally indistinguishable subspaces with dimension 4. We also explicitly construct locally indistinguishable subspaces with dimensions 3 and 5, respectively. Similar to the bipartite case, these results on multipartite locally indistinguishable subspaces can be used to estimate the one-shot environment-assisted classical capacity of a class of quantum broadcast channels.

关键词： Quantum computers

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：