基于知识表达的迁移学习研究
|
摘要
随着计算机信息技术的飞速发展，从海量数据中挖掘有用的信息并加以利用已经成为当前的研究热点。在数据挖掘中，传统的机器学习假设训练数据与测试数据具有相同的分布，而实际应用中往往并不能满足这一假设，即使勉强满足也大大降低了数据分析的效率。迁移学习方法能够有效地在相似的领域或任务之间进行信息的共享和迁移，使传统的从零开始的学习变成可积累的学习，具有成本低，效率高等优点，近几年成为数据挖掘和机器学习的关注焦点。
迁移学习最大的特点就是利用相关领域的知识来帮助完成目标领域中的学习任务。但是，不同的知识表达方式直接影响迁移学习的效果。本文从知识的表达方式入手，对知识表达与迁移方法的结合方式展开研究，主要工作如下：
首先，针对源领域数据和目标领域数据分布类似的情况，提出一种基于多源动态TrAdaBoost的实例迁移学习方法。该方法考虑多个源领域知识，使得目标任务的学习可以充分利用所有源领域信息；每次训练候选分类器时，所有源领域样本都参与学习，可以获得有利于目标任务学习的有用信息，从而避免负迁移的产生。理论分析验证了所提算法较单源迁移的优势，以及加入动态因子改善了源权重收敛导致的权重熵由源样本转移到目标样本的问题。实验结果表明，此算法在提高识别率和分类精度等方面的优势。
其次，针对源领域数据和目标领域数据分布差别较大的情况，提出一种基于协方差矩阵的特征迁移学习方法。该方法采用相似学习估计个体参数的协方差对，从数据间特征联系出发，通过构造半正定规划来组合这些估计值并学习当前任务的先验值，构建多变量高斯先验的协方差矩阵，进而对未标记的数据进行有效预测。实验结果表明，所提方法利用少量的源任务即可得到较好的迁移学习结果。
再次，针对共享知识是一些基函数的参数模型或先验分布的情况，提出一种基于层次贝叶斯的参数迁移学习方法。该方法定义任务相关性的层次概念，对标准Dirichlet过程模型进行扩展，建立多任务环境下的贝叶斯强化学习模型；有效学习类别数量、类型的推断过程以及用于探索新任务时的类别结构，从而更快地学习更多的知识。实验结果表明，提出的方法可以加快新领域中最优策略的收敛速度。
最后，针对共享知识是规则、结构和逻辑等关联规则的情况，提出一种基于马尔可夫逻辑网的关联规则迁移学习方法。该方法首先利用伪对数似然函数，将源领域中马尔可夫逻辑网表示的知识迁移到目标领域中，建立两个领域之间的关联；通过对源领域进行自诊断、结构更新和目标领域搜索新子句，来优化映射得到的结构，进而适应目标领域的学习。实验结果表明，算法成功地映射了迁移知识，提高了学习模型的精确度。
全文针对目前迁移学习研究中存在的部分问题，以如何利用源领域和目标领域间共享的不同知识表达方式为出发点，以提高目标领域的学习效率为目的，以不同的迁移方法为基本手段，对如何构造迁移环境和如何建立迁移环境下的有效迁移模型进行深入研究，提出了解决办法。实验结果验证了上述方法的可行性和有效性。论文研究成果丰富了迁移学习理论，为其在数据挖掘中的应用提供了有益的指导。
Along with the rapid development of computer information technology, to mineuseful information from mass data and put them into use have become a currentresearch hot spot. In data mining, the assumption for the traditional machine learningis that training data and test data have the same distribution. While in the practicalapplication, this assumption cannot be often met. To meet this assumption barely, theefficiency of data analysis has to be reduced. With the use of different fields ofknowledge for target task learning, transfer learning can transfer and share theinformation between similar domains or tasks, making the traditional learning fromscratch an addable one, which implies that the learning efficiency is higher and thecost is lower, so that it has become the focus in the field of data mining and machinelearning in recent years.
The most remarkable characteristic of transfer learning is that it can employ theknowledge in relative domains to help perform the learning tasks in the domain of thetarget. However, different ways of knowledge expression directly affect the effect oftransfer learning. Beginning with the ways of knowledge expression, aiming at thecombination modes of knowledge expression and transfer methods, this paper unfoldsits study mainly as follows:
Firstly, specific to the situation that the shared knowledge in the domains of thesource and the target are sample data with similar distribution, an instance transferlearning method based on multi-sources dynamic TrAdaBoost is put forward.Integrated with the knowledge in multiple source domains, this method makes thetarget task learning the one that is able to make good use of the information of allsource domains. Whenever candidate classifiers are trained, all the samples in allsource domains are involved in learning, and the information conducive to target tasklearning can be obtained, so that negative transfer can be avoided. The theoreticalanalysis suggests that the given algorithm better than single source transfer, by meansof adding the dynamic factor, this algorithm improves the defect that weight entropyto drift from source to target instances. The experimental results support that the givenalgorithm has the advantage of improving the recognition rate and classificationaccuracy.
Secondly, specific to the situation that shared knowledge in the domains of thesource and the target are sample data with different distribution, a feature transfer learning based on covariance matrix is proposed. This method employs similarlearning to estimate the covariance pairs of individual parameters. Starting from thecharacteristic relation between data, by means of constructing semidefinite program,the estimated values are combined, the priori values of the current tasks are emulated,and the priori multivariate gaussian covariance matrix is automatically built in order toeffectively predict the unlabelled data. The experimental results show that, with theuse of a small amount of the source tasks, good transfer learning results can beachieved by the given method.
Thirdly, specific to the situation that shared knowledge are the parametric modelsor the priori distributions of some base functions, a parameter transfer learningmethod based on hierarchical bayes is brought forward. With the use of this method,the hierarchy concept of task dependency is defined, the standard Dirichlet processmodel is developed, and then the bayes learning model is built under multi-taskenvironment. The number of categories, the inference process of types and thecategory structure used for exploring a new task are automatically studied, thus moreknowledge are learned quickly. The experimental results show that the proposedmethod can speed up the rate of convergence of the optimal strategy in the newdomain.
Finally, specific to the situation that shared knowledge are associated knowledgecontaining rules, structure and logic, an rule-based transfer learning methodassociated with Markov logic network is presented. Having applied this method, bymeans of pseudo log-likelihood function, the knowledge in the source domainexpressed in Markov logic network are transferred into the target domain while thelink between the two domains is established. By means of the self diagnosis andstructure update in the source domain and the new clause surf in the target domain,the mapped structure is optimized so that it can be adapt to the learning in the targetdomain. The experimental results show that the given algorithm successfully maps thetransferred knowledge, and improves the precision of the learning model.
Aiming at some existing problems in the current study on transfer learning, withhow to utilize the different knowledge expressions shared in both source and targetdomains as the starting point, with improving the learning efficiency in the targetdomain as the purpose, with different transfer methods as the basic means, with anin-depth study on how to construct the transfer environment and how to build theeffective transfer model under the transfer environment, some solutions are given in the full text. The related simulation results have verified the feasibility and availabilityof the given solutions. The research results in this paper have enriched transferlearning theories, which provide a helpful guidance for complex data mining.
迁移学习最大的特点就是利用相关领域的知识来帮助完成目标领域中的学习任务。但是，不同的知识表达方式直接影响迁移学习的效果。本文从知识的表达方式入手，对知识表达与迁移方法的结合方式展开研究，主要工作如下：
首先，针对源领域数据和目标领域数据分布类似的情况，提出一种基于多源动态TrAdaBoost的实例迁移学习方法。该方法考虑多个源领域知识，使得目标任务的学习可以充分利用所有源领域信息；每次训练候选分类器时，所有源领域样本都参与学习，可以获得有利于目标任务学习的有用信息，从而避免负迁移的产生。理论分析验证了所提算法较单源迁移的优势，以及加入动态因子改善了源权重收敛导致的权重熵由源样本转移到目标样本的问题。实验结果表明，此算法在提高识别率和分类精度等方面的优势。
其次，针对源领域数据和目标领域数据分布差别较大的情况，提出一种基于协方差矩阵的特征迁移学习方法。该方法采用相似学习估计个体参数的协方差对，从数据间特征联系出发，通过构造半正定规划来组合这些估计值并学习当前任务的先验值，构建多变量高斯先验的协方差矩阵，进而对未标记的数据进行有效预测。实验结果表明，所提方法利用少量的源任务即可得到较好的迁移学习结果。
再次，针对共享知识是一些基函数的参数模型或先验分布的情况，提出一种基于层次贝叶斯的参数迁移学习方法。该方法定义任务相关性的层次概念，对标准Dirichlet过程模型进行扩展，建立多任务环境下的贝叶斯强化学习模型；有效学习类别数量、类型的推断过程以及用于探索新任务时的类别结构，从而更快地学习更多的知识。实验结果表明，提出的方法可以加快新领域中最优策略的收敛速度。
最后，针对共享知识是规则、结构和逻辑等关联规则的情况，提出一种基于马尔可夫逻辑网的关联规则迁移学习方法。该方法首先利用伪对数似然函数，将源领域中马尔可夫逻辑网表示的知识迁移到目标领域中，建立两个领域之间的关联；通过对源领域进行自诊断、结构更新和目标领域搜索新子句，来优化映射得到的结构，进而适应目标领域的学习。实验结果表明，算法成功地映射了迁移知识，提高了学习模型的精确度。
全文针对目前迁移学习研究中存在的部分问题，以如何利用源领域和目标领域间共享的不同知识表达方式为出发点，以提高目标领域的学习效率为目的，以不同的迁移方法为基本手段，对如何构造迁移环境和如何建立迁移环境下的有效迁移模型进行深入研究，提出了解决办法。实验结果验证了上述方法的可行性和有效性。论文研究成果丰富了迁移学习理论，为其在数据挖掘中的应用提供了有益的指导。
Along with the rapid development of computer information technology, to mineuseful information from mass data and put them into use have become a currentresearch hot spot. In data mining, the assumption for the traditional machine learningis that training data and test data have the same distribution. While in the practicalapplication, this assumption cannot be often met. To meet this assumption barely, theefficiency of data analysis has to be reduced. With the use of different fields ofknowledge for target task learning, transfer learning can transfer and share theinformation between similar domains or tasks, making the traditional learning fromscratch an addable one, which implies that the learning efficiency is higher and thecost is lower, so that it has become the focus in the field of data mining and machinelearning in recent years.
The most remarkable characteristic of transfer learning is that it can employ theknowledge in relative domains to help perform the learning tasks in the domain of thetarget. However, different ways of knowledge expression directly affect the effect oftransfer learning. Beginning with the ways of knowledge expression, aiming at thecombination modes of knowledge expression and transfer methods, this paper unfoldsits study mainly as follows:
Firstly, specific to the situation that the shared knowledge in the domains of thesource and the target are sample data with similar distribution, an instance transferlearning method based on multi-sources dynamic TrAdaBoost is put forward.Integrated with the knowledge in multiple source domains, this method makes thetarget task learning the one that is able to make good use of the information of allsource domains. Whenever candidate classifiers are trained, all the samples in allsource domains are involved in learning, and the information conducive to target tasklearning can be obtained, so that negative transfer can be avoided. The theoreticalanalysis suggests that the given algorithm better than single source transfer, by meansof adding the dynamic factor, this algorithm improves the defect that weight entropyto drift from source to target instances. The experimental results support that the givenalgorithm has the advantage of improving the recognition rate and classificationaccuracy.
Secondly, specific to the situation that shared knowledge in the domains of thesource and the target are sample data with different distribution, a feature transfer learning based on covariance matrix is proposed. This method employs similarlearning to estimate the covariance pairs of individual parameters. Starting from thecharacteristic relation between data, by means of constructing semidefinite program,the estimated values are combined, the priori values of the current tasks are emulated,and the priori multivariate gaussian covariance matrix is automatically built in order toeffectively predict the unlabelled data. The experimental results show that, with theuse of a small amount of the source tasks, good transfer learning results can beachieved by the given method.
Thirdly, specific to the situation that shared knowledge are the parametric modelsor the priori distributions of some base functions, a parameter transfer learningmethod based on hierarchical bayes is brought forward. With the use of this method,the hierarchy concept of task dependency is defined, the standard Dirichlet processmodel is developed, and then the bayes learning model is built under multi-taskenvironment. The number of categories, the inference process of types and thecategory structure used for exploring a new task are automatically studied, thus moreknowledge are learned quickly. The experimental results show that the proposedmethod can speed up the rate of convergence of the optimal strategy in the newdomain.
Finally, specific to the situation that shared knowledge are associated knowledgecontaining rules, structure and logic, an rule-based transfer learning methodassociated with Markov logic network is presented. Having applied this method, bymeans of pseudo log-likelihood function, the knowledge in the source domainexpressed in Markov logic network are transferred into the target domain while thelink between the two domains is established. By means of the self diagnosis andstructure update in the source domain and the new clause surf in the target domain,the mapped structure is optimized so that it can be adapt to the learning in the targetdomain. The experimental results show that the given algorithm successfully maps thetransferred knowledge, and improves the precision of the learning model.
Aiming at some existing problems in the current study on transfer learning, withhow to utilize the different knowledge expressions shared in both source and targetdomains as the starting point, with improving the learning efficiency in the targetdomain as the purpose, with different transfer methods as the basic means, with anin-depth study on how to construct the transfer environment and how to build theeffective transfer model under the transfer environment, some solutions are given in the full text. The related simulation results have verified the feasibility and availabilityof the given solutions. The research results in this paper have enriched transferlearning theories, which provide a helpful guidance for complex data mining.
引文
[1] Frawley W.J., Piatetsky G., Shapiro C.. Knowledge discovery in databases: an overview[C].Piatetsky-Shapiro, Menlo Park, California: AAAI Press/The MIT Press,1991:1-27.
[2] Dietterich T. G., Domingos P., Getoor L., et al. Structured machine learning: the next ten years[J]. Machine Learning,2008,73(1):3-23.
[3]覃姜维.迁移学习方法研究及其在跨领域样本分类中的应用[D].广州:华南理工大学,2011.
[4]王皓,高阳,陈兴国.强化学习中的迁移：方法和进展[J].电子学报,2008,36(12):39-43.
[5] Yang Q.. An introduction to transfer learning [C]. Advanced Data Mining and ApplicationsConference, Proceedings4th International Conference, ADMA,2008:1-1.
[6] Matthew E.T., Peter S.. Transfer learning for reinforcement learning domains: a survey [J].Journal of Machine Learning,2009(10):1633-1685.
[7]孟佳娜.迁移学习在文本分类中的应用研究[D].大连:大连理工大学,2011.
[8]杜俊卫.基于聚类的文本迁移学习算法研究及应用[D].太原:山西财经大学,2011.
[9]方玉成.基于Logistic回归的直推式迁移学习方法研究[D].合肥:合肥工业大学,2010.
[10] Schmidhuber J.. On learning how to learn learning strategies. Technical Report FKI-198-94,Fakultat fur Informatik,1994:1-20.
[11] Thrun S.. Is learning the n-th thing any easier than learning the first?[J] Advances in NeuralInformation Processing Systems,1996:640-646.
[12] Caruana R.. Multitask learning [J]. Machine Learning,1997,28(1):41-75.
[13] Pan S., Yang Q.. A Survey on Transfer Learning [J]. IEEE Transactions on Knowledge andData Engineering,2009,22(10):1345-1359.
[14]戴文渊.基于实例和特征的迁移学习算法研究[D].上海:上海交通大学,2008.
[15] Ruckert U., Kramer S.. Kernel-based inductive transfer [C]. Machine Learning andKnowledge Discovery in Databases, European Conference. Heidelber: Springer-Verlag,2008:220-233.
[16] Arnold A., Nallapati R., Cohen W. W.. A comparative study of methods for transductiontransfer learning [C]. Proceedings of the7th IEEE International Conference on Data MiningWorkshops. Washington DC: IEEE,2007:77-82.
[17] Dai W., Xue G., Yang Q., et al. Transferring naive bayes classifiers for text classification [C].Proceedings of the22rd AAAI Conference on Artificial Intelligence. California: AAAI press,2007:540-545.
[18] Bickel S., Bruckner M., Scheffer T.. Discriminative learning for differing training and testdistributions [C]. Proceedings of the24th International Conference on Machine learning.New York: ACM,2007:81-88.
[19] Dai W., Yang Q., Xue G., et al. Boosting for transfer learning [C]. Proceedings of the24thInternational Conference on Machine Learning. New York: ACM,2007:193-200.
[20] Blitzer J., Dredze M., Pereira F.. Biographies, bollywood, boomboxes and blenders: Domainadaptation for sentiment classification [C]. Proceedings of the45th Annual Meeting of theAssociation of Computational Linguistics, Prague, Czech Republic,2007:432-439.
[21] Jebara T.. Multi-task feature and kernel selection for svms [C]. Proceedings of Twenty-FirstInternational Conference on Machine Learning,2004:433-440.
[22] Wang P., Domeniconi C.. Hu J.. Using wikipedia for co-clustering based cross-domain textclassification [C]. Proceedings of the8th IEEE International Converence on Data Mining.Wanshington DC: IEEE,2008:1085-1090.
[23] Argyriou A., Evgeniou T., Pontil M.. Multi-task feature learning [C]. Advances in NeuralInformation Processing Systems19,2007:41-48.
[24] Rajat R., Alexis B., Honglak L., et al. Self-taught learning: Transfer learning from unlabeleddata[C]. Proceedings of the24th International Conference on Machine Learning, Corvalis,Oregon, USA, June2007:759-766.
[25] Schwaighofer A., Tresp V., Yu K.. Learning gaussian process kernels via hierarchical bayes
[C]. Proceedings of the17th Annual Conference on Neural Information Processing Systems.Cambridge MIT Press,2005:1209-1216.
[26]王雪松,潘杰,程玉虎.基于图知识迁移的蚁群算法参数选择[J].控制与决策.2011,26(12):1855-1860.
[27] Evgeniou T., Pontil M.. Regularized multi-task learning [C]. Proceedings of the10thACMSIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM,2004:109-117.
[28] Gao J., Fan W., Jiang J., et al. Knowledge transfer via multiple model local structuremapping[C]. Proceedings of the14th ACM SIGKDD International Conference onKnowledge Discovery and Data Mining,2008:283-291.
[29] Davis J., Domingos P.. Deep transfer via second-order markov logic [C]. Proceedings of theAAAI-2008Workshop on Transfer Learning for Complex Tasks,2008:13-18.
[30] Kuncheva L.I., Rodrguez J. J.. Classifier ensembles with a random linear oracle [J]. IEEETransactions on Knowledge and Data Engineering,2007,19(4):500-508.
[31] Matthew R., Domingos P.. Markov logic networks [J]. Machine Learning Journal,2006,62:107-136.
[32]张倩,李明,王雪松.基于密度分布的半监督回归算法研究[J].工矿自动化.2012,38(3):29-30.
[33] Heckman J. J.. Sample selection bias as a specification error [J]. Economitrica,1979,47:153-161.
[34] Shimodaira H.. Improving predictive inference under covariate shift by weighting thelog-likelihood function [J]. Journal of Statistical Planning and Inference,2000,90(2):227-244.
[35] Zadrozny B.. Learning and evaluating classifiers under sample selection bias[C]. Proceedingsof the21st International Conference on Machine Learning. New York: ACM,2004:114-121.
[36] Fan W., Davidson I.. On sample selection bias and its efficient correction via model averagingand unlabeled examples[C]. Proceedings of the7th SIAM International Conference on DataMining. Philadelphia: SIAM,2007:320-331.
[37] Liao X., Xue Y., Carin L..Logistic regression with an auxiliary data source[C]. Proceedings ofthe21st International Conference on Machine Learning. New York: ACM,2005:505-512.
[38] Huang J., Smola A., Gretton A., et al. Correcting sample selection bias by unlabeled data[C].Proceedings of the19th Annual Conference on Neural Information Processing Systems.Cambridge: MIT Press,2007:601-608.
[39] Sugiyama M., Nakajima S., Kashima H., et al. Direct importance estimation with modelselection and its application to covariate shift adaptation[C]. Proceedings of the20thAnnualConference on Neural Information Processing Systems. Cambridge: MIT Press,2008:1433-1440.
[40] Wu P., Dietterich T. G.. Improving svm accuracy by training on auxiliary data sources [C].Proceedings of the21st International Conference on Machine Learning. New York: ACM,2004:110-117.
[41] Fan W., Davidson I., Zadrozny B., et al. An improved categorization of classifier's sensitivityon sample selection bias [C]. Proceedings of the5th IEEE International Conference on DataMining. Washington: IEEE,2005:605-608.
[42] Freund Y., Schapire R. E.. A decision-theoretic generalization of on-line learning and anapplication to boosting [J]. Journal of Computer and System Sciences,1997,55(1):119-139.
[43] Schapire R. E.. The strength of weak learnability [J]. Machine Learning,1990,5(2):197-227.
[44] Jiang J., Zhai C.. Instance weighting for domain adaptation in nlp[C]. Proceedings of the45thAnnual Meeting of the Association of Computational Linguistics. Belgium: Springer,2007:264-271.
[45] Bakker B., Heskes T.. Task clustering and gating for bayesian multitask learning [J]. Journalof Machine Learning Research,2003,4:83-99.
[46] Obozinski G., Taskar B., Jordan M.I.. Multi-task feature selection [J]. Technical report,Department of Statistics, University of California, Berkeley,2006:1-4.
[47] Ando R. K., Zhang T.. A high-performance semi-supervised learning method for textchunking [C]. Proceedings of the43rd Annual Meeting on Association for ComputationalLinguistics. ACL,2005:1-9.
[48] Blitzer J., McDonald R., Pereira F.. Domain adaptation with structural correspondencelearning [C]. Proceedings of the Conference on Empirical Methods in Natural Language.Sydney: EMNLP,2006:120-128.
[49] Micchelli C. A., Pontil M.. Kernels for multi-task learning [J]. Advances in NeuralInformation Processing Systems,2004:1-8.
[50] Evgeniou T., Micchelli C. A., Pontil M.. Learning multiple tasks with kernel methods [J].Journal of Machine Learning Research,2005,6:615-637.
[51] Yu K., Tresp V., Schwaighofer A.. Learning gaussian processes from multiple tasks [C].Proceedings of Twenty-Second International Conference on Machine Learning,2005:1017-1024.
[52] Dai W., Xue G., Yang Q., et al. Co-clustering based classification for out-of-domaindocuments[C]. Proceedings of the13th ACM SIGKDD International Conference onKnowledge Discovery and Data Mining. New York: ACM,2007:210-219.
[53] Li B., Yang Q., Xue X.. Can movies and books collaborate cross-domain collaborativefiltering for sparsity reduction[C]. Proceedings of the20thInternational Joint Conference onArtificial Interlligence.San Francisco: Morgan Kaufmann Publishers,2009:2025-2057.
[54] Pan S. J., Kwok J. T., Yang Q.. Transfer learning via dimensionality reduction[C].Proceedings of the23th AAAI Conference on Artificial Intelligence. California: AAAI press.2008:677-682.
[55] Pan S. J., Tsang I. W., Kwok J. T,. et al. Domain adaptation via transfer component analysis[J].IEEE Transactions on Neural Network,2011,22(2):199-210.
[56] Pan S. J., Shen D., Yang Q., et al. Transferring localization models across space[C].Proceedings of the23rd AAAI Conference on Artificial Intelligence.California: AAAI press,2008:1383-1388.
[57] Boyd S., Vandenberghe L.. Convex optimization [M]. Cambridge University Press,2004.
[58] Zhong E., Fan W., Peng J., et al. Cross domain distribution adaptation via kernel mapping[C].Proceedings of the15th ACM SIGKDD international conference on Knowledge discoveryand data mining. New York: ACM,2009:1027-1035.
[59] Ren J., Shi X., Fan W., et al. Type-independent correction of sample selection bias viastructural discovery and re-balancing[C]. Proceedings of the8th SIAM InternationalConference on Data Mining. Philadelphia: SDM,2008:565-576.
[60] Ling X., Dai W., Xue G., et al. Spectral domain transfer learning[C]. Proceedings of the14thACM SIGKDD International Conference on Knowledge Discovery and Data Mining. NewYork: ACM,2008:488-496.
[61] Shi X., Yang Q., Ren J.. Relaxed transfer of different classes via spectral partition [C].Machine Learning and Knowledge Discovery in Databases, European Conference. Belgium:Springer,2009:366-381.
[62] He J., Liu Y., Lawrence R.. Graph based transfer learning[C]. Proceeding of the18thACMconference on Information and knowledge management. New York: ACM,2009:937-946.
[63] Dai W., Jin O., Xue G. R., et al. Eigen transfer: a unified framework for transfer learning [C].Proceedings of the26th International Conference on Machine Learning. NewYork: ACM,2009:193-200.
[64]朱美强,程玉虎,李明等.一类基于谱方法的强化学习混合迁移算法[J].自动化学报,2012.38(11):1665-1676.
[65] Dai W., Yang Q., Xue G., et al. Self-taught clustering[C]. Proceedings of the25thInternationalConference of Machine Learning. New York: ACM,2008:200-207.
[66] Quattoni A., Collins M., Darrell T.. Transfer learning for image classification with sparseprototype representation [C]. IEEE Conference on Computer Vision and Pattern Recognition.Wanshington DC: IEEE,2008:1-8.
[67] Su B., Xu W., Shen Y.. Margin-based transfer learning [C]. Proceedings of he SixthInternational Symposium on Neural Networks. Belgium: Springer,2009:223-233.
[68] Daume III H., Marcu D.. Domain adaptation for statistical classifiers [J]. Journal of ArtificialIntelligence Research,2006,26(1):101-126.
[69] Lawrence N. D., Platt J. C.. Learning to learn with the informative vector machine [C].Proceedings of the21st International Conference on Machine Learning. New York: ACM,2004:512-519.
[70] Bonilla E., Chai K. M., Williams C.. Multi-task gaussian process prediction [C]. Proceedingsof the20th Annual Conference on Neural Information Processing Systems. Cambridge: MITPress,2008:153-160.
[71] Raina R., Ng A.Y., Koller D.. Constructing informative priors using transfer learning[C].Proceedings of the23rd International Conference on Machine Learning, Pittsburgh,Pennsylvania, USA, June2006:713-720.
[72] Ling X., Xue G.R., Dai W., et al. Can chinese web pages be classified with English datasource?[C]. Proceedings of the17th International Conference on World Wide Web. Beijing,China: ACM, April2008:969-978.
[73] Steffen B., Tobias S.. Dirichlet-enhanced spam filtering based on biased samples[C].Advances in Neural Information Processing Systems,2007:161-168.
[74] Zhang X., Dai W., Xue G., et al. Adaptive email spam filtering based on informationtheory[C]. Proceedings of the Eighth International Conference on Web Information SystemsEngineering,2007:159-170.
[75] Raykar V.C., Krishnapuram B., Bi J.,et al. Bayesian multiple instance learning: automaticfeature selection and inductive transfer[C]. Proceedings of the25thInternational Conferenceon Machine learning. Helsinki, Finland: ACM, July2008:808-815.
[76] Liu Y., Peter S.. Value-function-based transfer for reinforcement learning using structuremapping [C]. Proceedings of the Twenty-First National Conference on Artificial Intelligence,2007:415-420.
[77] Pan S.J., Kwok J.T., Yang Q.,et al. Adaptive localization in a dynamic WiFi environmentthrough multi-view learning [C]. Proceedings of the22nd AAAI Conference on ArtificialIntelligence, Vancouver, British Columbia, Canada, July2007:1108-1113.
[78] Zheng V W, Xiang E W, Yang Q, et al. Transferring localization models over time.Proceedings of the23th AAAI Conference on Artificial Intelligence [C]. California: AAAIpress,2008:1421-1426.
[79] Zhuo H., Yang Q., Hu D.H., et al. Transferring knowledge from another domain for learningaction models[C]. Proceedings of10th Pacific Rim International Conference on ArtificialIntelligence, December2008:1110-1115.
[80]王岚.基于人工神经网络的学习模型迁移方法的实现[C].第十届全球华人计算机教育应用会议论文集.2006:155-160.
[81] Fung G.P.C., Yu J.X., Lu H., et al. Text classification without negative examples revisit [J].IEEE Transactions on Knowledge and Data Engineering,2006,18(1):6-20.
[82] Al-Mubaid H., Umair S. A.. A new text categorization technique using distributionalclustering and learning logic [J]. IEEE Transactions on Knowledge and Data Engineering,2006,18(9):1156-1165.
[83] Pan S.J., Zheng V.W., Yang Q., et al. Transfer learning for wifi-based indoor localization[C].Proceedings of the Workshop on Transfer Learning for Complex Task of the23rdAAAIConference on Artificial Intelligence,Chicago,Illinois,USA,July,2008:43-48.
[84] http://www.ecmlpkdd2006.org/challenge.html
[85] http://www.cse.ust.hk/~qyang/ICDMDMC2007
[86] http://www.cis.upenn.edu/~mdredze/datasets/sentiment
[87]陈德品.基于迁移学习的跨领域排序学习算法研究[D].合肥:中国科学技术大学,2010.
[88] Thomas G.. Dietterich. Ensemble learning [J]. In The Handbook of Brain Theory and NeuralNetworks, Second Edition,2002:1-9.
[89]梁延峰.基于专家委员会的主动学习算法研究[D].青岛:中国海洋大学,2010.
[90]刘志刚,李德仁,秦前清.支持向量机在多类分类问题中的推广[J].计算机工程与应用,2004,22(10):10-13.
[91]葛启发,冯夏庭.基于AdaBoost组合学习方法的岩爆分类预测研究[J]..岩土力学,2008,29(4):943-948.
[92]许志杰.迁移学习理论与算法研究[D].上海:华东师范大学,2012.
[93] Crammer, K., Kearns, M., Wortman, J.. Learning from multiple sources [J]. Journal ofMachine Learning Research,2008,13(9):1757-1774.
[94] Michael T.R., Zvika M., Leslie P. K.. To transfer or not to transfer [R]. A NIPS-05WorkshopOil Inductive Transfer,2005:1-4.
[95] Yang J., YanR., Hauptmann A.G.. Cross-domain video concept detection using adaptiveSVMs [C]. Proceedings of the ACM International Multimedia Conference and Exhibition,2007:188-197.
[96] Jiang W., Zavesky E., Chang S.F., et al. Cross-domain learning methods for high-level visualconcept classifcation[C].15th IEEE International Conference on Image Processing,2008:161-164.
[97] Duan L., Tsang I. W., Xu D., Maybank S. J.. Domain transfer SVM for video conceptdetection[C]. IEEE Computer Society Conference on Computer Vision and PatternRecognition Workshops,2009:1375-1381.
[98] Lampert C. H., NickischH., Harmeling S.. Learning to detect unseen object class by betweenclass attibute transfer [C]. IEEE Computer Society Conference on Computer Vision andPattern Recognition Workshops,2009:951-958.
[99] Torrey L., Shavlik J.. Transfer learning[C]. Handbook of research on machine learningapplications, published by IGI Global,2009:1-22.
[100] Pardoe, D., Stone, P. Boosting for regression transfer[C]. Proceedings of the27thInternational Conference on Machine Learning,2010:863-870.
[101] Eaton, E., desJardins, M. Set-based boosting for instance-level transfer r[C]. Proceedings ofthe2009IEEE International Conference on Data Mining Workshops,2009:422-428.
[102] Eaton, E. Selective knowledge transfer for machine learning [D]. University of MarylandBaltimore County,2009.
[103] Yao, Y., Doretto, G. Boosting for transfer learning with multiple sources [C]. Proceedings ofthe IEEE Conference on Computer Vision and Pattern Recognition,2010:1855-1862.
[104] Venkatesan, A., Krishnan, N., Panchanathan, S. Cost-sensitive boosting for concept drift [C].Proceedings of the2010International Workshop on Handling Concept Drift in AdaptiveInformation Systems,2010:41-47.
[105] Sun, Y., Kamel, M.S., Wong, A.K., Wang, Y. Cost-sensitive boosting for classification ofimbalanced data [J]. Pattern Recognition,2007:40(12):3358-3378.
[106] Littlestone, N., Warmuth, M.K. The weighted majority algorithm [J]. InformationComputation.1994,108:256-261.
[107] Samir A.S., Chandan K. R.. Adaptive boosting for transfer learning using dynamicupdates[C]. Machine Learning and Knowledge Discovery in Databases-EuropeanConference,2011:60-75.
[108] Vapnik V. N.. Estimation of dependences based on empirical data [M]. Springer-Verlag,1982.
[109] Griffn G., Holub A., Perona P.. Caltech-256object category dataset [R]. Technical Report7694, California Institute of Technology,2007.
[110] Lee S.I., Chatalbashev V., Viekrey D., et al. Learning a meta-level prior for feature relevancefrom multiple related tasks[C]. Proceeding of the24thinternational conference on Machinelearning. Corvalis, Oregon: ACM,2007,489-496.
[111] Wang C. Mahadevan S.. Manifold alignment using procrustes analysis[C]. Proceedings ofthe25thinternational conference on Machine learning. Helsinki, inland: ACM,2008:1120-1127.
[112]范东凯.摄像机自标定的算法研究[D].哈尔滨:哈尔滨工业大学,2010.
[113]柴晶.雷达高分辨距离像目标识别的拒判算法和特征提取技术研究[D].西安:西安电子科技大学,2010.
[114]张杰.基于协方差矩阵的稳健目标匹配算法研究[D].秦皇岛:燕山大学,2012.
[115]孙琦.基于协方差矩阵的目标跟踪方法[D].黑龙江:黑龙江大学,2012.
[116] Zhang Q., Li M., Wang X.S.. Global Path Planning Method of Mobile Robot in UncertainEnvironment [C]. Proceedings of2010Chinese Control and Decision,2010:4320-4324.
[117]宁忠磊,王宏琦,张正.一种基于协方差矩阵的自动目标检测方法[J].中国科学院研究生院学报.2010,27(3):370-375.
[118] Boyd S., Vandenberghe L.. Convex optimization[M]. Cambridge University Press,2004.
[119] Sturm J. Using Sedumil, a Matlab toolbox for optimization over symmetric cones [J].Optimization Methods and Software.1999,11(1):625-653.
[120] Nigam, K., Lafferty, J., McCallum, A.Using maximum entropy for text classification [C].IJCAI Workshop on Machine Learning for Information Filtering.1999:1-7.
[121]刘赫.文本分类中若干问题研究[D].长春:吉林大学,2009.
[122] Xie S., Fan S., Peng J. et al.. Latent space domain transfer between high dimensionaloverlapping distribution [C]. Proceeding of the18thInternational Conference on World WildWeb,2009:91-100.
[123] http://sedumi.mcmaster.ca
[124] Ng, A. Y., Jordan, M., Weiss, Y.. On spectral clustering: Analysis and an algorithm[C]. NIPS,2002:1-8.
[125] http://infomap.stanford.edu
[126] Miller, G. A.. Wordnet: A lexical database for English [J]. Commun. ACM,1995:38,39-41.
[127] http://dmoz.org
[128]蔡伟宏.基于非参数贝叶斯方法的资产配置[D].武汉:华中科技大学,2012.
[129]武晓嘉,杨盘洪.数据融合技术在农业专家系统中的应用[J].太原理工大学学报,2004:13-15.
[130] Refik S., Minje S.. Bayesian analysis of nonhomogeneous markov chains[C]. InternationalSociety for Bayesian Analysis,2005:61-65.
[131]朱志勇,刘伟铭,伍友龙.基于Byaes法则和BP神经网络的高速动态情形下车型识别[J].计算机测量与控制,2005:101-102.
[132] Paul J.M., Jacobus J.J., Andriette B.. A Bayesian method to analyse cancer survival timesusing the weibull model[C]. International Society for Bayesian Analysis,2005:78-81.
[l33]王利民.贝叶斯学习理论中若干问题的研究[D].吉林:吉林大学,2005.
[134] Bishop C.M. Pattern recognition and machine learning [M], Springer,2006.
[135] Tipping.M.. Bayesian inference: an introduction to principles and practice in machineLearning. Advanced Lectures on Machine Leaining,2004,31(76):41-62.
[136] Ioannis N., Athanassios K., Dimitris K.. Bayesian assessment of the distribution of insuranceclaim counts using reversible jump MCMC[J]. North American Actuarial Journal,2005,9(11):1-12.
[137] Ntzoufras I.. Bayesian modeling of outstanding liabilities incorporating claim countuncertainty [J]. North American Actuarial Journal2002.6(1):113-128.
[138]宫秀军.贝叶斯学习理论及其应用研究[D].合肥:中国科学院研究生院,2002.
[139]曲寒冰.贝叶斯框架下广义结合泛函网络与广义约束神经网络的研究与应用[D].合肥:中国科学院研究生院,2007.
[140]贾乃光译.统计决策论及贝叶斯分析[M].北京:中国统计出版社,1998.
[141] Kotz S.,吴喜之.现代贝叶斯统计学.[M].北京:中国统计出版社,2000.
[142] Schnieper.R. Robust Bayesian experience rating[J]. Astin Bulletin,2004,34(1):125-150.
[143]王晓园.贝叶斯方法在保险精算中的应用研究[D].重庆:重庆理工大学.2011.
[144]王雪松,潘杰,程玉虎.基于知识迁移的Ant-Q算法[J].电子学报,2011,39(10):2359-2365.
[145] Jaynes E.T.. Probability theory: The logic of science [M]. New York: Cambridge UniversityPress,2003.
[146]陈东.基于层次贝叶斯的子空间分类[D].杭州:浙江大学,2008.
[147] Ng A.Y., Jordan M.I.. Approximate inference algorithms for two-layer Bayesian networks.Advances in Neural Information Processing Systems12and Cambridge MA:MIT Press,2000:1-7.
[148] Jordan M.I., Ghahramani Z., Jaakkola T.S.. An introduction to variational methods forgraphical models[J]. Machine Learning,1999,37(2):183-233.
[149]任周鹏.三峡库区蓄水前后鼠密度空间分布统计推断[D].长春:东北师范大学,2011
[150] Lawson A.. Bayesian disease mapping: hierarchical modeling in spatial epidemiology[M].Chapman and Hall/CRC,2008.
[151] Konidaris G., Barto A. Autonomous shaping: knowledge transfer in reinforcementlearning[C]. Proceedings of the23rdinternational conference on machine learning,2006,489-496.
[152] Banerjee B., Stone P.. General game learning using knowledge transfer[C]. Proceedings ofthe20th international joint conference on artificial intelligence,2007:1-6.
[153] Mehta N., Natarajan S., Tadepalli, P., et al. Transfer in variable-reward hierarchicalreinforcement learning [J]. Machine Learning,2008,73(3):289-312.
[154] Zhang Q., Li M., Wang X.S., et al. Reinforcement Learning in Robot Path Optimization [J].Journal of Software,2012,7(3):657-662.
[155] Sutton R. S., Barto A. G.. Reinforcement learning: an introduction [M]. Cambridge, MA:MIT Press,1998.
[156] Duff M.. Design for an optimal probe[C]. International conference on machine learning,2003:131-138.
[157] Zhang Q., Li M.. The Knowledge Transfer Methods and Its Application in Robot Based onSubtask Hierarchical Reinforcement Learning [J]. Advances in Information Sciences andService Sciences,2012,4(20):119-128.
[158] Neal R. M. Markov chain sampling methods for dirichlet process mixture models[J]. Journalof Computational and Graphical Statistics,2000,9(2),249-265.
[159] Sullivan H.L., Curtin C.G., Reynolds C.A., et al. The effect of topography on the foragingcosts of heteromyid rodents [J]. Journal of Arid Environments,2001,48(3):255-266.
[160] Mandelik Y., Jones M., Dayan T. Structurally complex habitat and sensory adaptationsmediate the behavioral responses of a desert rodent to an indirect cue for increased predationrisk[J]. Evolutionary Ecology Research,2003:501-515.
[161]朱明慧,韩玉启.基于正态-逆Wishart先验分布的贝叶斯分类识别方法研究[J].工程数学学报.2004,1:53-56.
[162] Domingos P., Richardson M.. Markov logic: A unifying framework for statistical relationallearning[C]. Proceedings of the ICML-2004Workshop on Statistical Relational Learning andits Connections to Other Fields. Banff,2004:49-54.
[163] Koller D., Friedman N.. Probabilistic graphical models: principles and techniques [M].Cambridge: The MIT Press,2009.
[164]徐从富,郝春亮,苏保君等.马尔可夫逻辑网络研究[J].软件学报,2011,22(8):1699-1713.
[165] Domingos P., Kok S., Poon H., et al. Unifying logical and statistical AI [C]. Proceedings ofthe21st National conference on Artificial Intelligence. Boston,2006:2-7.
[166] Domingos P.. What's missing in AI: The interface layer[C]. Proceedings of the ArtificialIntelligence: The1st Hundred Years. Menlo Park: AAAI Press,2006:1-6.
[167] Kok S., Domingos P.. Statistical predicate invention [C]. Proceedings of the24th conferenceon Machine Learning. Corvallis,2007:433-440.
[168] Singla P., Domingos P.. Discriminative training of Markov logic networks[C]. Proceedingsof the20th National conference on Artificial Intelligence. Pittsburgh,2005:868-873.
[169] Kok S., Domingos P.. Learning the structure of Markov logic networks[C]. Proceedings ofthe22nd Conference on Machine Learning Bonn,2005:441-448.
[170] Wang J., Domingos P.. Hybrid Markov logic networks[C]. Proceedings of the23rd NationalConference on Artificial Intelligence. Chicago,2008:1106-1111.
[171] Poon H., Domingos P.. Unsupervised semantic parsing [C]. Proceedings of the2009conference on Empirical Methods in Natural Language Processing. Singapore,2009:1-10.
[172] Kok S., Domingos P.. Learning Markov logic networks using structural motifs[C].Proceedings of the27th Conference on Machine Learning. Haifa,2010:551-558.
[173]孔润.马尔可夫逻辑网在超文本分类与链接预测中的应用[D].重庆:重庆大学,2011.
[174]黄涛.马尔可夫逻辑网在Web中的应用[D].重庆:重庆大学,2010.
[175]耿素云,屈婉玲.离散数学[M].北京:高等教育出版社,1998.
[176] Russell S,. Norvig P.. Artificial intelligence: a modern approach[J]. Upper Saddle River:Pearson Education,2003,16(3):73.
[177]孙舒杨.统计关系学习的若干问题研究[D].吉林:吉林大学,2006.
[178] Lowd D., Domingos P.. Efficient weight learning for Markov logic networks[C].Proceedings of the eleventh European conference on principles and practice of knowledgediscovery in databases.Warsaw, Poland: Springer,2007:200-211.
[179]熊忠阳,舒方俊,张玉芳等. Markov逻辑网在迁移学习中的应用[J].计算机工程,2011,37(24):58-160.
[180] Wasserman S., Faust K.. Social network analysis: Methods and applications[M].Cambridge,UK:Cambridge University Press,1994.
[181] Poon H., Domingos P.. Sound and effcient inference with probabilistic and deterministicdependencies[C]. Proceedings of the Twenty-First National Conference on ArtifcialIntelligence. Boston, MA,2006:458-463.
[182] Whittaker J.. Graphical models in applied multivariate statistics[M].John Wiley and Sons,1990.
[183] Richards B. L., Mooney R. J.. Learning relations by pathfnding[C]. AAAI’92, Proceedingsof the tenth national conference on Artificial intelligence. Springer,1992:50-55.
[184] Mihalkova L S.. Learning with Markov logic networks transfer learning, structure learning,and an application to Web query disambiguation [D]. The University of Texas At Austin.Doctor of Philosophy,2009.
[185] Mihalkova L. S.. Improving learning of Markov logic networks using transfer andbottom-up induction[D]. Texas, USA: Department of Computer Sciences University of Texasat Austin, University of Texas at Austin,2009.
[186] Mihalkova L., Mooney R.J.. Bottom-Up learning of Markov logic network structure [C].Proceedings of the24th Conference on Machine Learning. Corvalis,2007:625-632.
[187] Richards, B. L., Mooney, R. J.. Automated refnement of frst-order Horn-clause domaintheories[J]. Machine Learning,1995,19(2):95-131.
[188] Kok S., Domingos P.. Learning Markov logic network structure via hypergraph lifting [C].Proceedings of the26th conference on Machine Learning. Montreal,2009:505-512.
[189] IMDB Dataset. http://www.imdb.com.
[190] UW-CSE Dataset. http://www.cs.washington.edu/ai/mln/database.html.
[191] Craven M., DiPasquo D., Freitag D.,et al.. Learning to extract symbolic knowledge from theWorld Wide Web [C]. Proceedings of the Fifteenth National Conference on ArtifcialIntelligence, Springer,1998:509-516.
[192] Kok S, Singla P, Richardson M, et al. The Alchemy System for Statistical Relational AI[Z].Department of Computer Science and Engineering, University of Washington,2005.
[193] Cohen, P. R., Chang, Y., Morrison, C. T.. Learning and transferring action schema[C].Proceedings of the Twentieth International Joint Conference on Artifcial Intelligence.Hyderabad, India,2007:720-725.
[2] Dietterich T. G., Domingos P., Getoor L., et al. Structured machine learning: the next ten years[J]. Machine Learning,2008,73(1):3-23.
[3]覃姜维.迁移学习方法研究及其在跨领域样本分类中的应用[D].广州:华南理工大学,2011.
[4]王皓,高阳,陈兴国.强化学习中的迁移：方法和进展[J].电子学报,2008,36(12):39-43.
[5] Yang Q.. An introduction to transfer learning [C]. Advanced Data Mining and ApplicationsConference, Proceedings4th International Conference, ADMA,2008:1-1.
[6] Matthew E.T., Peter S.. Transfer learning for reinforcement learning domains: a survey [J].Journal of Machine Learning,2009(10):1633-1685.
[7]孟佳娜.迁移学习在文本分类中的应用研究[D].大连:大连理工大学,2011.
[8]杜俊卫.基于聚类的文本迁移学习算法研究及应用[D].太原:山西财经大学,2011.
[9]方玉成.基于Logistic回归的直推式迁移学习方法研究[D].合肥:合肥工业大学,2010.
[10] Schmidhuber J.. On learning how to learn learning strategies. Technical Report FKI-198-94,Fakultat fur Informatik,1994:1-20.
[11] Thrun S.. Is learning the n-th thing any easier than learning the first?[J] Advances in NeuralInformation Processing Systems,1996:640-646.
[12] Caruana R.. Multitask learning [J]. Machine Learning,1997,28(1):41-75.
[13] Pan S., Yang Q.. A Survey on Transfer Learning [J]. IEEE Transactions on Knowledge andData Engineering,2009,22(10):1345-1359.
[14]戴文渊.基于实例和特征的迁移学习算法研究[D].上海:上海交通大学,2008.
[15] Ruckert U., Kramer S.. Kernel-based inductive transfer [C]. Machine Learning andKnowledge Discovery in Databases, European Conference. Heidelber: Springer-Verlag,2008:220-233.
[16] Arnold A., Nallapati R., Cohen W. W.. A comparative study of methods for transductiontransfer learning [C]. Proceedings of the7th IEEE International Conference on Data MiningWorkshops. Washington DC: IEEE,2007:77-82.
[17] Dai W., Xue G., Yang Q., et al. Transferring naive bayes classifiers for text classification [C].Proceedings of the22rd AAAI Conference on Artificial Intelligence. California: AAAI press,2007:540-545.
[18] Bickel S., Bruckner M., Scheffer T.. Discriminative learning for differing training and testdistributions [C]. Proceedings of the24th International Conference on Machine learning.New York: ACM,2007:81-88.
[19] Dai W., Yang Q., Xue G., et al. Boosting for transfer learning [C]. Proceedings of the24thInternational Conference on Machine Learning. New York: ACM,2007:193-200.
[20] Blitzer J., Dredze M., Pereira F.. Biographies, bollywood, boomboxes and blenders: Domainadaptation for sentiment classification [C]. Proceedings of the45th Annual Meeting of theAssociation of Computational Linguistics, Prague, Czech Republic,2007:432-439.
[21] Jebara T.. Multi-task feature and kernel selection for svms [C]. Proceedings of Twenty-FirstInternational Conference on Machine Learning,2004:433-440.
[22] Wang P., Domeniconi C.. Hu J.. Using wikipedia for co-clustering based cross-domain textclassification [C]. Proceedings of the8th IEEE International Converence on Data Mining.Wanshington DC: IEEE,2008:1085-1090.
[23] Argyriou A., Evgeniou T., Pontil M.. Multi-task feature learning [C]. Advances in NeuralInformation Processing Systems19,2007:41-48.
[24] Rajat R., Alexis B., Honglak L., et al. Self-taught learning: Transfer learning from unlabeleddata[C]. Proceedings of the24th International Conference on Machine Learning, Corvalis,Oregon, USA, June2007:759-766.
[25] Schwaighofer A., Tresp V., Yu K.. Learning gaussian process kernels via hierarchical bayes
[C]. Proceedings of the17th Annual Conference on Neural Information Processing Systems.Cambridge MIT Press,2005:1209-1216.
[26]王雪松,潘杰,程玉虎.基于图知识迁移的蚁群算法参数选择[J].控制与决策.2011,26(12):1855-1860.
[27] Evgeniou T., Pontil M.. Regularized multi-task learning [C]. Proceedings of the10thACMSIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM,2004:109-117.
[28] Gao J., Fan W., Jiang J., et al. Knowledge transfer via multiple model local structuremapping[C]. Proceedings of the14th ACM SIGKDD International Conference onKnowledge Discovery and Data Mining,2008:283-291.
[29] Davis J., Domingos P.. Deep transfer via second-order markov logic [C]. Proceedings of theAAAI-2008Workshop on Transfer Learning for Complex Tasks,2008:13-18.
[30] Kuncheva L.I., Rodrguez J. J.. Classifier ensembles with a random linear oracle [J]. IEEETransactions on Knowledge and Data Engineering,2007,19(4):500-508.
[31] Matthew R., Domingos P.. Markov logic networks [J]. Machine Learning Journal,2006,62:107-136.
[32]张倩,李明,王雪松.基于密度分布的半监督回归算法研究[J].工矿自动化.2012,38(3):29-30.
[33] Heckman J. J.. Sample selection bias as a specification error [J]. Economitrica,1979,47:153-161.
[34] Shimodaira H.. Improving predictive inference under covariate shift by weighting thelog-likelihood function [J]. Journal of Statistical Planning and Inference,2000,90(2):227-244.
[35] Zadrozny B.. Learning and evaluating classifiers under sample selection bias[C]. Proceedingsof the21st International Conference on Machine Learning. New York: ACM,2004:114-121.
[36] Fan W., Davidson I.. On sample selection bias and its efficient correction via model averagingand unlabeled examples[C]. Proceedings of the7th SIAM International Conference on DataMining. Philadelphia: SIAM,2007:320-331.
[37] Liao X., Xue Y., Carin L..Logistic regression with an auxiliary data source[C]. Proceedings ofthe21st International Conference on Machine Learning. New York: ACM,2005:505-512.
[38] Huang J., Smola A., Gretton A., et al. Correcting sample selection bias by unlabeled data[C].Proceedings of the19th Annual Conference on Neural Information Processing Systems.Cambridge: MIT Press,2007:601-608.
[39] Sugiyama M., Nakajima S., Kashima H., et al. Direct importance estimation with modelselection and its application to covariate shift adaptation[C]. Proceedings of the20thAnnualConference on Neural Information Processing Systems. Cambridge: MIT Press,2008:1433-1440.
[40] Wu P., Dietterich T. G.. Improving svm accuracy by training on auxiliary data sources [C].Proceedings of the21st International Conference on Machine Learning. New York: ACM,2004:110-117.
[41] Fan W., Davidson I., Zadrozny B., et al. An improved categorization of classifier's sensitivityon sample selection bias [C]. Proceedings of the5th IEEE International Conference on DataMining. Washington: IEEE,2005:605-608.
[42] Freund Y., Schapire R. E.. A decision-theoretic generalization of on-line learning and anapplication to boosting [J]. Journal of Computer and System Sciences,1997,55(1):119-139.
[43] Schapire R. E.. The strength of weak learnability [J]. Machine Learning,1990,5(2):197-227.
[44] Jiang J., Zhai C.. Instance weighting for domain adaptation in nlp[C]. Proceedings of the45thAnnual Meeting of the Association of Computational Linguistics. Belgium: Springer,2007:264-271.
[45] Bakker B., Heskes T.. Task clustering and gating for bayesian multitask learning [J]. Journalof Machine Learning Research,2003,4:83-99.
[46] Obozinski G., Taskar B., Jordan M.I.. Multi-task feature selection [J]. Technical report,Department of Statistics, University of California, Berkeley,2006:1-4.
[47] Ando R. K., Zhang T.. A high-performance semi-supervised learning method for textchunking [C]. Proceedings of the43rd Annual Meeting on Association for ComputationalLinguistics. ACL,2005:1-9.
[48] Blitzer J., McDonald R., Pereira F.. Domain adaptation with structural correspondencelearning [C]. Proceedings of the Conference on Empirical Methods in Natural Language.Sydney: EMNLP,2006:120-128.
[49] Micchelli C. A., Pontil M.. Kernels for multi-task learning [J]. Advances in NeuralInformation Processing Systems,2004:1-8.
[50] Evgeniou T., Micchelli C. A., Pontil M.. Learning multiple tasks with kernel methods [J].Journal of Machine Learning Research,2005,6:615-637.
[51] Yu K., Tresp V., Schwaighofer A.. Learning gaussian processes from multiple tasks [C].Proceedings of Twenty-Second International Conference on Machine Learning,2005:1017-1024.
[52] Dai W., Xue G., Yang Q., et al. Co-clustering based classification for out-of-domaindocuments[C]. Proceedings of the13th ACM SIGKDD International Conference onKnowledge Discovery and Data Mining. New York: ACM,2007:210-219.
[53] Li B., Yang Q., Xue X.. Can movies and books collaborate cross-domain collaborativefiltering for sparsity reduction[C]. Proceedings of the20thInternational Joint Conference onArtificial Interlligence.San Francisco: Morgan Kaufmann Publishers,2009:2025-2057.
[54] Pan S. J., Kwok J. T., Yang Q.. Transfer learning via dimensionality reduction[C].Proceedings of the23th AAAI Conference on Artificial Intelligence. California: AAAI press.2008:677-682.
[55] Pan S. J., Tsang I. W., Kwok J. T,. et al. Domain adaptation via transfer component analysis[J].IEEE Transactions on Neural Network,2011,22(2):199-210.
[56] Pan S. J., Shen D., Yang Q., et al. Transferring localization models across space[C].Proceedings of the23rd AAAI Conference on Artificial Intelligence.California: AAAI press,2008:1383-1388.
[57] Boyd S., Vandenberghe L.. Convex optimization [M]. Cambridge University Press,2004.
[58] Zhong E., Fan W., Peng J., et al. Cross domain distribution adaptation via kernel mapping[C].Proceedings of the15th ACM SIGKDD international conference on Knowledge discoveryand data mining. New York: ACM,2009:1027-1035.
[59] Ren J., Shi X., Fan W., et al. Type-independent correction of sample selection bias viastructural discovery and re-balancing[C]. Proceedings of the8th SIAM InternationalConference on Data Mining. Philadelphia: SDM,2008:565-576.
[60] Ling X., Dai W., Xue G., et al. Spectral domain transfer learning[C]. Proceedings of the14thACM SIGKDD International Conference on Knowledge Discovery and Data Mining. NewYork: ACM,2008:488-496.
[61] Shi X., Yang Q., Ren J.. Relaxed transfer of different classes via spectral partition [C].Machine Learning and Knowledge Discovery in Databases, European Conference. Belgium:Springer,2009:366-381.
[62] He J., Liu Y., Lawrence R.. Graph based transfer learning[C]. Proceeding of the18thACMconference on Information and knowledge management. New York: ACM,2009:937-946.
[63] Dai W., Jin O., Xue G. R., et al. Eigen transfer: a unified framework for transfer learning [C].Proceedings of the26th International Conference on Machine Learning. NewYork: ACM,2009:193-200.
[64]朱美强,程玉虎,李明等.一类基于谱方法的强化学习混合迁移算法[J].自动化学报,2012.38(11):1665-1676.
[65] Dai W., Yang Q., Xue G., et al. Self-taught clustering[C]. Proceedings of the25thInternationalConference of Machine Learning. New York: ACM,2008:200-207.
[66] Quattoni A., Collins M., Darrell T.. Transfer learning for image classification with sparseprototype representation [C]. IEEE Conference on Computer Vision and Pattern Recognition.Wanshington DC: IEEE,2008:1-8.
[67] Su B., Xu W., Shen Y.. Margin-based transfer learning [C]. Proceedings of he SixthInternational Symposium on Neural Networks. Belgium: Springer,2009:223-233.
[68] Daume III H., Marcu D.. Domain adaptation for statistical classifiers [J]. Journal of ArtificialIntelligence Research,2006,26(1):101-126.
[69] Lawrence N. D., Platt J. C.. Learning to learn with the informative vector machine [C].Proceedings of the21st International Conference on Machine Learning. New York: ACM,2004:512-519.
[70] Bonilla E., Chai K. M., Williams C.. Multi-task gaussian process prediction [C]. Proceedingsof the20th Annual Conference on Neural Information Processing Systems. Cambridge: MITPress,2008:153-160.
[71] Raina R., Ng A.Y., Koller D.. Constructing informative priors using transfer learning[C].Proceedings of the23rd International Conference on Machine Learning, Pittsburgh,Pennsylvania, USA, June2006:713-720.
[72] Ling X., Xue G.R., Dai W., et al. Can chinese web pages be classified with English datasource?[C]. Proceedings of the17th International Conference on World Wide Web. Beijing,China: ACM, April2008:969-978.
[73] Steffen B., Tobias S.. Dirichlet-enhanced spam filtering based on biased samples[C].Advances in Neural Information Processing Systems,2007:161-168.
[74] Zhang X., Dai W., Xue G., et al. Adaptive email spam filtering based on informationtheory[C]. Proceedings of the Eighth International Conference on Web Information SystemsEngineering,2007:159-170.
[75] Raykar V.C., Krishnapuram B., Bi J.,et al. Bayesian multiple instance learning: automaticfeature selection and inductive transfer[C]. Proceedings of the25thInternational Conferenceon Machine learning. Helsinki, Finland: ACM, July2008:808-815.
[76] Liu Y., Peter S.. Value-function-based transfer for reinforcement learning using structuremapping [C]. Proceedings of the Twenty-First National Conference on Artificial Intelligence,2007:415-420.
[77] Pan S.J., Kwok J.T., Yang Q.,et al. Adaptive localization in a dynamic WiFi environmentthrough multi-view learning [C]. Proceedings of the22nd AAAI Conference on ArtificialIntelligence, Vancouver, British Columbia, Canada, July2007:1108-1113.
[78] Zheng V W, Xiang E W, Yang Q, et al. Transferring localization models over time.Proceedings of the23th AAAI Conference on Artificial Intelligence [C]. California: AAAIpress,2008:1421-1426.
[79] Zhuo H., Yang Q., Hu D.H., et al. Transferring knowledge from another domain for learningaction models[C]. Proceedings of10th Pacific Rim International Conference on ArtificialIntelligence, December2008:1110-1115.
[80]王岚.基于人工神经网络的学习模型迁移方法的实现[C].第十届全球华人计算机教育应用会议论文集.2006:155-160.
[81] Fung G.P.C., Yu J.X., Lu H., et al. Text classification without negative examples revisit [J].IEEE Transactions on Knowledge and Data Engineering,2006,18(1):6-20.
[82] Al-Mubaid H., Umair S. A.. A new text categorization technique using distributionalclustering and learning logic [J]. IEEE Transactions on Knowledge and Data Engineering,2006,18(9):1156-1165.
[83] Pan S.J., Zheng V.W., Yang Q., et al. Transfer learning for wifi-based indoor localization[C].Proceedings of the Workshop on Transfer Learning for Complex Task of the23rdAAAIConference on Artificial Intelligence,Chicago,Illinois,USA,July,2008:43-48.
[84] http://www.ecmlpkdd2006.org/challenge.html
[85] http://www.cse.ust.hk/~qyang/ICDMDMC2007
[86] http://www.cis.upenn.edu/~mdredze/datasets/sentiment
[87]陈德品.基于迁移学习的跨领域排序学习算法研究[D].合肥:中国科学技术大学,2010.
[88] Thomas G.. Dietterich. Ensemble learning [J]. In The Handbook of Brain Theory and NeuralNetworks, Second Edition,2002:1-9.
[89]梁延峰.基于专家委员会的主动学习算法研究[D].青岛:中国海洋大学,2010.
[90]刘志刚,李德仁,秦前清.支持向量机在多类分类问题中的推广[J].计算机工程与应用,2004,22(10):10-13.
[91]葛启发,冯夏庭.基于AdaBoost组合学习方法的岩爆分类预测研究[J]..岩土力学,2008,29(4):943-948.
[92]许志杰.迁移学习理论与算法研究[D].上海:华东师范大学,2012.
[93] Crammer, K., Kearns, M., Wortman, J.. Learning from multiple sources [J]. Journal ofMachine Learning Research,2008,13(9):1757-1774.
[94] Michael T.R., Zvika M., Leslie P. K.. To transfer or not to transfer [R]. A NIPS-05WorkshopOil Inductive Transfer,2005:1-4.
[95] Yang J., YanR., Hauptmann A.G.. Cross-domain video concept detection using adaptiveSVMs [C]. Proceedings of the ACM International Multimedia Conference and Exhibition,2007:188-197.
[96] Jiang W., Zavesky E., Chang S.F., et al. Cross-domain learning methods for high-level visualconcept classifcation[C].15th IEEE International Conference on Image Processing,2008:161-164.
[97] Duan L., Tsang I. W., Xu D., Maybank S. J.. Domain transfer SVM for video conceptdetection[C]. IEEE Computer Society Conference on Computer Vision and PatternRecognition Workshops,2009:1375-1381.
[98] Lampert C. H., NickischH., Harmeling S.. Learning to detect unseen object class by betweenclass attibute transfer [C]. IEEE Computer Society Conference on Computer Vision andPattern Recognition Workshops,2009:951-958.
[99] Torrey L., Shavlik J.. Transfer learning[C]. Handbook of research on machine learningapplications, published by IGI Global,2009:1-22.
[100] Pardoe, D., Stone, P. Boosting for regression transfer[C]. Proceedings of the27thInternational Conference on Machine Learning,2010:863-870.
[101] Eaton, E., desJardins, M. Set-based boosting for instance-level transfer r[C]. Proceedings ofthe2009IEEE International Conference on Data Mining Workshops,2009:422-428.
[102] Eaton, E. Selective knowledge transfer for machine learning [D]. University of MarylandBaltimore County,2009.
[103] Yao, Y., Doretto, G. Boosting for transfer learning with multiple sources [C]. Proceedings ofthe IEEE Conference on Computer Vision and Pattern Recognition,2010:1855-1862.
[104] Venkatesan, A., Krishnan, N., Panchanathan, S. Cost-sensitive boosting for concept drift [C].Proceedings of the2010International Workshop on Handling Concept Drift in AdaptiveInformation Systems,2010:41-47.
[105] Sun, Y., Kamel, M.S., Wong, A.K., Wang, Y. Cost-sensitive boosting for classification ofimbalanced data [J]. Pattern Recognition,2007:40(12):3358-3378.
[106] Littlestone, N., Warmuth, M.K. The weighted majority algorithm [J]. InformationComputation.1994,108:256-261.
[107] Samir A.S., Chandan K. R.. Adaptive boosting for transfer learning using dynamicupdates[C]. Machine Learning and Knowledge Discovery in Databases-EuropeanConference,2011:60-75.
[108] Vapnik V. N.. Estimation of dependences based on empirical data [M]. Springer-Verlag,1982.
[109] Griffn G., Holub A., Perona P.. Caltech-256object category dataset [R]. Technical Report7694, California Institute of Technology,2007.
[110] Lee S.I., Chatalbashev V., Viekrey D., et al. Learning a meta-level prior for feature relevancefrom multiple related tasks[C]. Proceeding of the24thinternational conference on Machinelearning. Corvalis, Oregon: ACM,2007,489-496.
[111] Wang C. Mahadevan S.. Manifold alignment using procrustes analysis[C]. Proceedings ofthe25thinternational conference on Machine learning. Helsinki, inland: ACM,2008:1120-1127.
[112]范东凯.摄像机自标定的算法研究[D].哈尔滨:哈尔滨工业大学,2010.
[113]柴晶.雷达高分辨距离像目标识别的拒判算法和特征提取技术研究[D].西安:西安电子科技大学,2010.
[114]张杰.基于协方差矩阵的稳健目标匹配算法研究[D].秦皇岛:燕山大学,2012.
[115]孙琦.基于协方差矩阵的目标跟踪方法[D].黑龙江:黑龙江大学,2012.
[116] Zhang Q., Li M., Wang X.S.. Global Path Planning Method of Mobile Robot in UncertainEnvironment [C]. Proceedings of2010Chinese Control and Decision,2010:4320-4324.
[117]宁忠磊,王宏琦,张正.一种基于协方差矩阵的自动目标检测方法[J].中国科学院研究生院学报.2010,27(3):370-375.
[118] Boyd S., Vandenberghe L.. Convex optimization[M]. Cambridge University Press,2004.
[119] Sturm J. Using Sedumil, a Matlab toolbox for optimization over symmetric cones [J].Optimization Methods and Software.1999,11(1):625-653.
[120] Nigam, K., Lafferty, J., McCallum, A.Using maximum entropy for text classification [C].IJCAI Workshop on Machine Learning for Information Filtering.1999:1-7.
[121]刘赫.文本分类中若干问题研究[D].长春:吉林大学,2009.
[122] Xie S., Fan S., Peng J. et al.. Latent space domain transfer between high dimensionaloverlapping distribution [C]. Proceeding of the18thInternational Conference on World WildWeb,2009:91-100.
[123] http://sedumi.mcmaster.ca
[124] Ng, A. Y., Jordan, M., Weiss, Y.. On spectral clustering: Analysis and an algorithm[C]. NIPS,2002:1-8.
[125] http://infomap.stanford.edu
[126] Miller, G. A.. Wordnet: A lexical database for English [J]. Commun. ACM,1995:38,39-41.
[127] http://dmoz.org
[128]蔡伟宏.基于非参数贝叶斯方法的资产配置[D].武汉:华中科技大学,2012.
[129]武晓嘉,杨盘洪.数据融合技术在农业专家系统中的应用[J].太原理工大学学报,2004:13-15.
[130] Refik S., Minje S.. Bayesian analysis of nonhomogeneous markov chains[C]. InternationalSociety for Bayesian Analysis,2005:61-65.
[131]朱志勇,刘伟铭,伍友龙.基于Byaes法则和BP神经网络的高速动态情形下车型识别[J].计算机测量与控制,2005:101-102.
[132] Paul J.M., Jacobus J.J., Andriette B.. A Bayesian method to analyse cancer survival timesusing the weibull model[C]. International Society for Bayesian Analysis,2005:78-81.
[l33]王利民.贝叶斯学习理论中若干问题的研究[D].吉林:吉林大学,2005.
[134] Bishop C.M. Pattern recognition and machine learning [M], Springer,2006.
[135] Tipping.M.. Bayesian inference: an introduction to principles and practice in machineLearning. Advanced Lectures on Machine Leaining,2004,31(76):41-62.
[136] Ioannis N., Athanassios K., Dimitris K.. Bayesian assessment of the distribution of insuranceclaim counts using reversible jump MCMC[J]. North American Actuarial Journal,2005,9(11):1-12.
[137] Ntzoufras I.. Bayesian modeling of outstanding liabilities incorporating claim countuncertainty [J]. North American Actuarial Journal2002.6(1):113-128.
[138]宫秀军.贝叶斯学习理论及其应用研究[D].合肥:中国科学院研究生院,2002.
[139]曲寒冰.贝叶斯框架下广义结合泛函网络与广义约束神经网络的研究与应用[D].合肥:中国科学院研究生院,2007.
[140]贾乃光译.统计决策论及贝叶斯分析[M].北京:中国统计出版社,1998.
[141] Kotz S.,吴喜之.现代贝叶斯统计学.[M].北京:中国统计出版社,2000.
[142] Schnieper.R. Robust Bayesian experience rating[J]. Astin Bulletin,2004,34(1):125-150.
[143]王晓园.贝叶斯方法在保险精算中的应用研究[D].重庆:重庆理工大学.2011.
[144]王雪松,潘杰,程玉虎.基于知识迁移的Ant-Q算法[J].电子学报,2011,39(10):2359-2365.
[145] Jaynes E.T.. Probability theory: The logic of science [M]. New York: Cambridge UniversityPress,2003.
[146]陈东.基于层次贝叶斯的子空间分类[D].杭州:浙江大学,2008.
[147] Ng A.Y., Jordan M.I.. Approximate inference algorithms for two-layer Bayesian networks.Advances in Neural Information Processing Systems12and Cambridge MA:MIT Press,2000:1-7.
[148] Jordan M.I., Ghahramani Z., Jaakkola T.S.. An introduction to variational methods forgraphical models[J]. Machine Learning,1999,37(2):183-233.
[149]任周鹏.三峡库区蓄水前后鼠密度空间分布统计推断[D].长春:东北师范大学,2011
[150] Lawson A.. Bayesian disease mapping: hierarchical modeling in spatial epidemiology[M].Chapman and Hall/CRC,2008.
[151] Konidaris G., Barto A. Autonomous shaping: knowledge transfer in reinforcementlearning[C]. Proceedings of the23rdinternational conference on machine learning,2006,489-496.
[152] Banerjee B., Stone P.. General game learning using knowledge transfer[C]. Proceedings ofthe20th international joint conference on artificial intelligence,2007:1-6.
[153] Mehta N., Natarajan S., Tadepalli, P., et al. Transfer in variable-reward hierarchicalreinforcement learning [J]. Machine Learning,2008,73(3):289-312.
[154] Zhang Q., Li M., Wang X.S., et al. Reinforcement Learning in Robot Path Optimization [J].Journal of Software,2012,7(3):657-662.
[155] Sutton R. S., Barto A. G.. Reinforcement learning: an introduction [M]. Cambridge, MA:MIT Press,1998.
[156] Duff M.. Design for an optimal probe[C]. International conference on machine learning,2003:131-138.
[157] Zhang Q., Li M.. The Knowledge Transfer Methods and Its Application in Robot Based onSubtask Hierarchical Reinforcement Learning [J]. Advances in Information Sciences andService Sciences,2012,4(20):119-128.
[158] Neal R. M. Markov chain sampling methods for dirichlet process mixture models[J]. Journalof Computational and Graphical Statistics,2000,9(2),249-265.
[159] Sullivan H.L., Curtin C.G., Reynolds C.A., et al. The effect of topography on the foragingcosts of heteromyid rodents [J]. Journal of Arid Environments,2001,48(3):255-266.
[160] Mandelik Y., Jones M., Dayan T. Structurally complex habitat and sensory adaptationsmediate the behavioral responses of a desert rodent to an indirect cue for increased predationrisk[J]. Evolutionary Ecology Research,2003:501-515.
[161]朱明慧,韩玉启.基于正态-逆Wishart先验分布的贝叶斯分类识别方法研究[J].工程数学学报.2004,1:53-56.
[162] Domingos P., Richardson M.. Markov logic: A unifying framework for statistical relationallearning[C]. Proceedings of the ICML-2004Workshop on Statistical Relational Learning andits Connections to Other Fields. Banff,2004:49-54.
[163] Koller D., Friedman N.. Probabilistic graphical models: principles and techniques [M].Cambridge: The MIT Press,2009.
[164]徐从富,郝春亮,苏保君等.马尔可夫逻辑网络研究[J].软件学报,2011,22(8):1699-1713.
[165] Domingos P., Kok S., Poon H., et al. Unifying logical and statistical AI [C]. Proceedings ofthe21st National conference on Artificial Intelligence. Boston,2006:2-7.
[166] Domingos P.. What's missing in AI: The interface layer[C]. Proceedings of the ArtificialIntelligence: The1st Hundred Years. Menlo Park: AAAI Press,2006:1-6.
[167] Kok S., Domingos P.. Statistical predicate invention [C]. Proceedings of the24th conferenceon Machine Learning. Corvallis,2007:433-440.
[168] Singla P., Domingos P.. Discriminative training of Markov logic networks[C]. Proceedingsof the20th National conference on Artificial Intelligence. Pittsburgh,2005:868-873.
[169] Kok S., Domingos P.. Learning the structure of Markov logic networks[C]. Proceedings ofthe22nd Conference on Machine Learning Bonn,2005:441-448.
[170] Wang J., Domingos P.. Hybrid Markov logic networks[C]. Proceedings of the23rd NationalConference on Artificial Intelligence. Chicago,2008:1106-1111.
[171] Poon H., Domingos P.. Unsupervised semantic parsing [C]. Proceedings of the2009conference on Empirical Methods in Natural Language Processing. Singapore,2009:1-10.
[172] Kok S., Domingos P.. Learning Markov logic networks using structural motifs[C].Proceedings of the27th Conference on Machine Learning. Haifa,2010:551-558.
[173]孔润.马尔可夫逻辑网在超文本分类与链接预测中的应用[D].重庆:重庆大学,2011.
[174]黄涛.马尔可夫逻辑网在Web中的应用[D].重庆:重庆大学,2010.
[175]耿素云,屈婉玲.离散数学[M].北京:高等教育出版社,1998.
[176] Russell S,. Norvig P.. Artificial intelligence: a modern approach[J]. Upper Saddle River:Pearson Education,2003,16(3):73.
[177]孙舒杨.统计关系学习的若干问题研究[D].吉林:吉林大学,2006.
[178] Lowd D., Domingos P.. Efficient weight learning for Markov logic networks[C].Proceedings of the eleventh European conference on principles and practice of knowledgediscovery in databases.Warsaw, Poland: Springer,2007:200-211.
[179]熊忠阳,舒方俊,张玉芳等. Markov逻辑网在迁移学习中的应用[J].计算机工程,2011,37(24):58-160.
[180] Wasserman S., Faust K.. Social network analysis: Methods and applications[M].Cambridge,UK:Cambridge University Press,1994.
[181] Poon H., Domingos P.. Sound and effcient inference with probabilistic and deterministicdependencies[C]. Proceedings of the Twenty-First National Conference on ArtifcialIntelligence. Boston, MA,2006:458-463.
[182] Whittaker J.. Graphical models in applied multivariate statistics[M].John Wiley and Sons,1990.
[183] Richards B. L., Mooney R. J.. Learning relations by pathfnding[C]. AAAI’92, Proceedingsof the tenth national conference on Artificial intelligence. Springer,1992:50-55.
[184] Mihalkova L S.. Learning with Markov logic networks transfer learning, structure learning,and an application to Web query disambiguation [D]. The University of Texas At Austin.Doctor of Philosophy,2009.
[185] Mihalkova L. S.. Improving learning of Markov logic networks using transfer andbottom-up induction[D]. Texas, USA: Department of Computer Sciences University of Texasat Austin, University of Texas at Austin,2009.
[186] Mihalkova L., Mooney R.J.. Bottom-Up learning of Markov logic network structure [C].Proceedings of the24th Conference on Machine Learning. Corvalis,2007:625-632.
[187] Richards, B. L., Mooney, R. J.. Automated refnement of frst-order Horn-clause domaintheories[J]. Machine Learning,1995,19(2):95-131.
[188] Kok S., Domingos P.. Learning Markov logic network structure via hypergraph lifting [C].Proceedings of the26th conference on Machine Learning. Montreal,2009:505-512.
[189] IMDB Dataset. http://www.imdb.com.
[190] UW-CSE Dataset. http://www.cs.washington.edu/ai/mln/database.html.
[191] Craven M., DiPasquo D., Freitag D.,et al.. Learning to extract symbolic knowledge from theWorld Wide Web [C]. Proceedings of the Fifteenth National Conference on ArtifcialIntelligence, Springer,1998:509-516.
[192] Kok S, Singla P, Richardson M, et al. The Alchemy System for Statistical Relational AI[Z].Department of Computer Science and Engineering, University of Washington,2005.
[193] Cohen, P. R., Chang, Y., Morrison, C. T.. Learning and transferring action schema[C].Proceedings of the Twentieth International Joint Conference on Artifcial Intelligence.Hyderabad, India,2007:720-725.