深度学习技术的发展使得基于深度神经网络的方法成为自然语言处理(Natural Language Processing,NLP)领域的一种新解决思路。虽然神经网络技术能够有效提升生成文本的质量,但是生成文本的内容很容易偏离作者原本要表达的语义。笔者按照...深度学习技术的发展使得基于深度神经网络的方法成为自然语言处理(Natural Language Processing,NLP)领域的一种新解决思路。虽然神经网络技术能够有效提升生成文本的质量,但是生成文本的内容很容易偏离作者原本要表达的语义。笔者按照人们的写作习惯,即先构思出各部分的主题再进行写作的方式,提出基于交叉项编码的关键词主题控制文本生成模型。与其他模型相比,该模型生成的句子不仅在双语互译质量评估(Bilingual Evaluation Understudy,BLEU)中的得分更高,Correlation值也要高于其他模型。展开更多
To adjust the variance of source rate in linear broadcast networks, global encoding kernels should have corresponding dimensions to instruct the decoding process. The algorithm of constructing such global encoding ker...To adjust the variance of source rate in linear broadcast networks, global encoding kernels should have corresponding dimensions to instruct the decoding process. The algorithm of constructing such global encoding kernels is to adjust heterogeneous network to possible link failures. Linear algebra, graph theory and group theory are applied to construct one series of global encoding kernels which are applicable to all source rates. The effectiveness and existence of such global encoding kernels are proved. Based on 2 information flow, the algorithm of construction is explicitly given within polynomial time O（｜E｜ ｜T｜.ω^2max）, and the memory complexity of algorithm is O（｜E｜）. Both time and memory complexity of this algorithm proposed can be O（ωmax） less than those of algorithms in related works.展开更多
In bi-directional three-node cooperation, one regenerative strategy with network coding and power optimization is proposed for system sum-rate under a total energy constraint. In this paper, the network coding and pow...In bi-directional three-node cooperation, one regenerative strategy with network coding and power optimization is proposed for system sum-rate under a total energy constraint. In this paper, the network coding and power optimization are applied to improve system sum-rate. But max-rain optimization problem in power allocation is a NP-hard problem. In high Signal-to-Noise Ratio regime, this NP-hard problem is transformed into constrained polynomial optimization problem, which can be computed in polynomial time. Although it is a suboptimal solution, numerical simulations show that this strategy enhances the system sum-rate up to 45% as compared to a traditional four-phase strategy, and up to 13% as compared to the three-phase strategy without power optimization.展开更多
The famous F5 algorithm for computing Grobner basis was presented by Faugere in 2002. The original version of F5 is given in programming codes, so it is a bit difficult to understand. In this paper, the F5 algorithm i...The famous F5 algorithm for computing Grobner basis was presented by Faugere in 2002. The original version of F5 is given in programming codes, so it is a bit difficult to understand. In this paper, the F5 algorithm is simplified as F5B in a Buchberger＇s style such that it is easy to understand and implement. In order to describe F5B, we introduce F5-reduction, which keeps the signature of labeled polynomials unchanged after reduction. The equivalence between F5 and F5B is also shown. At last, some versions of the F5 algorithm are illustrated.展开更多
文摘深度学习技术的发展使得基于深度神经网络的方法成为自然语言处理(Natural Language Processing,NLP)领域的一种新解决思路。虽然神经网络技术能够有效提升生成文本的质量,但是生成文本的内容很容易偏离作者原本要表达的语义。笔者按照人们的写作习惯,即先构思出各部分的主题再进行写作的方式,提出基于交叉项编码的关键词主题控制文本生成模型。与其他模型相比,该模型生成的句子不仅在双语互译质量评估(Bilingual Evaluation Understudy,BLEU)中的得分更高,Correlation值也要高于其他模型。
基金Project(60872005) supported by National Natural Science Foundation of China
文摘To adjust the variance of source rate in linear broadcast networks, global encoding kernels should have corresponding dimensions to instruct the decoding process. The algorithm of constructing such global encoding kernels is to adjust heterogeneous network to possible link failures. Linear algebra, graph theory and group theory are applied to construct one series of global encoding kernels which are applicable to all source rates. The effectiveness and existence of such global encoding kernels are proved. Based on 2 information flow, the algorithm of construction is explicitly given within polynomial time O（｜E｜ ｜T｜.ω^2max）, and the memory complexity of algorithm is O（｜E｜）. Both time and memory complexity of this algorithm proposed can be O（ωmax） less than those of algorithms in related works.
基金Supported by the High Technology Research and Development Program of China (No. 2006AA01Z282 2007CB310608)
文摘In bi-directional three-node cooperation, one regenerative strategy with network coding and power optimization is proposed for system sum-rate under a total energy constraint. In this paper, the network coding and power optimization are applied to improve system sum-rate. But max-rain optimization problem in power allocation is a NP-hard problem. In high Signal-to-Noise Ratio regime, this NP-hard problem is transformed into constrained polynomial optimization problem, which can be computed in polynomial time. Although it is a suboptimal solution, numerical simulations show that this strategy enhances the system sum-rate up to 45% as compared to a traditional four-phase strategy, and up to 13% as compared to the three-phase strategy without power optimization.
文摘The famous F5 algorithm for computing Grobner basis was presented by Faugere in 2002. The original version of F5 is given in programming codes, so it is a bit difficult to understand. In this paper, the F5 algorithm is simplified as F5B in a Buchberger＇s style such that it is easy to understand and implement. In order to describe F5B, we introduce F5-reduction, which keeps the signature of labeled polynomials unchanged after reduction. The equivalence between F5 and F5B is also shown. At last, some versions of the F5 algorithm are illustrated.
