基于终身学习Agent的多源迁移算法研究
|
摘要
终身学习Agent在智能优化、机器学习、模式识别与图像处理等领域面临路径寻优、文本分类、人脸识别、色彩增强与最优决策等一系列问题时，不可避免地会遭遇连续空间维数灾、目标训练样本匮乏获取代价高以及多次重复面临相似任务等情形，本文针对终身学习Agent的上述特点，采用如下研究方法实现其在不同领域内的多源迁移学习：
1.机器学习领域中预测分类在训练样本较少时会出现精度下降的问题，为此提出一种多源决策树自适应迁移方法。首先，自适应地采用成分预测概率或路径预测概率对决策树间的相似性进行判定，其次，根据多源判定条件确定是否采用多源集成迁移。同时考虑目标训练样本降低到极端情形，即仅有唯一样本时的单样本人脸识别问题。提出一种基于LPP特征映射的多源迁移算法，并采用FERET、ORL与Yale等典型的人脸识别数据库进行识别验证。
2.强化学习在面临大尺度或连续空间复杂系统时遭遇维数灾难题，提出一种基于极限学习机（Extreme Learning Machine，ELM）的多源迁移Q学习算法，ELM采用神经网络映射机制保证Q值函数的逼近，而多源迁移机制能够降低目标问题的决策难度。迁移的本质在于任务空间与样本空间的相似度衡量，利用先验概率尽可能地确保迁移的任务与样本能够在目标任务中起到积极的作用，尽量避免负迁移的发生。
3.图像处理领域中由于色彩序列模糊性与不确定性造成的色彩扭曲问题，提出一种基于主动轮廓探索的多源色彩迁移算法。利用主动进化方法生成虚拟轮廓线，并采用能量函数评价机制迫使虚拟轮廓线逐渐逼近实际轮廓线。同时考虑源图像与目标图像在RGB、Gray和LMS等不同色彩空间的表示、分割、转换，实现其在l空间的多源色彩迁移。单源与多源色彩迁移的对比、灰度化色彩通道的选择等实验验证了所提算法的合理性与有效性。
4.智能优化算法具有随问题规模指数级增长的计算复杂度以及对自身多变量耦合参数设置的依赖性，为此分别提出多源迁移Ant-Q学习算法与基于图构建的多源参数迁移算法。前者通过贝叶斯理论分析源任务与目标任务的相似率，并以此为权值确定各源任务的迁移样本；后者则是将包含知识（蚁群算法运行参数）的源任务构造模型迁移图，以逼近多变量参数的流形空间，并进一步扩充模型迁移图实现源任务参数到目标任务参数的映射。
For problems such as path planning, text classification, face recognition, colorenhancement and optimal decisions, lifelong learning Agent will inevitably encountermassive data processing, deficient target training samples, high costs and multiplyrepeated tasks in intelligent optimization, machine learning, pattern recognition andimage processing. Regarding the above features of lifelong learning Agent, the papertries to apply the following research methods to achieve multi-source transfer learningin different fields:
1. In machine learning, lack of training samples in classification prediction can leadto accuracy drop, hence, MSTDT method is proposed. At first, it will determine thesimilarity among decision trees by automatically selecting component probability orpath probability; secondly, it can choose whether to use multi-source integratedtransfer based on multi-source conditions. At the same time an extremely low targettraining samples or only one available sample shall be taken into account to analyzeface identification. At last it puts forward multi-source transfer algorithm based onLPP characteristic mapping and verifies the identification by using typical faceidentification database like FERET, ORL and Yale.
2. An ELM multi-source transfer Q learning algorithm is brought forward whenreinforcement learning faces large scale or continuous complex curse ofdimensionality problems. ELM ensures the approximation of Q value function, whilethe multi-source transfer mechanism can reduce decision difficulty of target problems.In fact, the nature of transfer is the similarity measurement between task space andsample space, and by using prior probability one can ensure transfer task and sampleplay a positive role in the targeted task and prevent negative transfer from occurring.
3. In color processing, due to color distortion caused by color sequence ambiguityand uncertainty, the paper proposes a multi-source color transfer algorithm based onactive profile exploration. It uses active evolution methods to generate virtual contourand applies energy function evaluation mechanism to force it is gradually approachingactual contour. Meanwhile consideration is taken for the expression, split andconversion of source and target images in different color spaces such as RGB, Grayand LMS to achieve its multi-source color transfer in l space. The comparisonand gray color channel selection tests of single and multi-source transfer prove thereasonability and effectiveness of the algorithm.
4. Intelligent optimized algorithm has different computational complexity withexponential growth and dependence on its multivariable coupling parameter settings,so the paper proposes Multi-Source Transfer Ant-Q and multi-source parametertransfer algorithm based on graph construction. The former analyzes the similarityratio between source and target tasks and determine each transfer samples by thisweight; the later constructs the model transfer graph of the source task includingknowledge (ACO operating parameters) to approximate the manifold space ofmultivariate parameters.
1.机器学习领域中预测分类在训练样本较少时会出现精度下降的问题，为此提出一种多源决策树自适应迁移方法。首先，自适应地采用成分预测概率或路径预测概率对决策树间的相似性进行判定，其次，根据多源判定条件确定是否采用多源集成迁移。同时考虑目标训练样本降低到极端情形，即仅有唯一样本时的单样本人脸识别问题。提出一种基于LPP特征映射的多源迁移算法，并采用FERET、ORL与Yale等典型的人脸识别数据库进行识别验证。
2.强化学习在面临大尺度或连续空间复杂系统时遭遇维数灾难题，提出一种基于极限学习机（Extreme Learning Machine，ELM）的多源迁移Q学习算法，ELM采用神经网络映射机制保证Q值函数的逼近，而多源迁移机制能够降低目标问题的决策难度。迁移的本质在于任务空间与样本空间的相似度衡量，利用先验概率尽可能地确保迁移的任务与样本能够在目标任务中起到积极的作用，尽量避免负迁移的发生。
3.图像处理领域中由于色彩序列模糊性与不确定性造成的色彩扭曲问题，提出一种基于主动轮廓探索的多源色彩迁移算法。利用主动进化方法生成虚拟轮廓线，并采用能量函数评价机制迫使虚拟轮廓线逐渐逼近实际轮廓线。同时考虑源图像与目标图像在RGB、Gray和LMS等不同色彩空间的表示、分割、转换，实现其在l空间的多源色彩迁移。单源与多源色彩迁移的对比、灰度化色彩通道的选择等实验验证了所提算法的合理性与有效性。
4.智能优化算法具有随问题规模指数级增长的计算复杂度以及对自身多变量耦合参数设置的依赖性，为此分别提出多源迁移Ant-Q学习算法与基于图构建的多源参数迁移算法。前者通过贝叶斯理论分析源任务与目标任务的相似率，并以此为权值确定各源任务的迁移样本；后者则是将包含知识（蚁群算法运行参数）的源任务构造模型迁移图，以逼近多变量参数的流形空间，并进一步扩充模型迁移图实现源任务参数到目标任务参数的映射。
For problems such as path planning, text classification, face recognition, colorenhancement and optimal decisions, lifelong learning Agent will inevitably encountermassive data processing, deficient target training samples, high costs and multiplyrepeated tasks in intelligent optimization, machine learning, pattern recognition andimage processing. Regarding the above features of lifelong learning Agent, the papertries to apply the following research methods to achieve multi-source transfer learningin different fields:
1. In machine learning, lack of training samples in classification prediction can leadto accuracy drop, hence, MSTDT method is proposed. At first, it will determine thesimilarity among decision trees by automatically selecting component probability orpath probability; secondly, it can choose whether to use multi-source integratedtransfer based on multi-source conditions. At the same time an extremely low targettraining samples or only one available sample shall be taken into account to analyzeface identification. At last it puts forward multi-source transfer algorithm based onLPP characteristic mapping and verifies the identification by using typical faceidentification database like FERET, ORL and Yale.
2. An ELM multi-source transfer Q learning algorithm is brought forward whenreinforcement learning faces large scale or continuous complex curse ofdimensionality problems. ELM ensures the approximation of Q value function, whilethe multi-source transfer mechanism can reduce decision difficulty of target problems.In fact, the nature of transfer is the similarity measurement between task space andsample space, and by using prior probability one can ensure transfer task and sampleplay a positive role in the targeted task and prevent negative transfer from occurring.
3. In color processing, due to color distortion caused by color sequence ambiguityand uncertainty, the paper proposes a multi-source color transfer algorithm based onactive profile exploration. It uses active evolution methods to generate virtual contourand applies energy function evaluation mechanism to force it is gradually approachingactual contour. Meanwhile consideration is taken for the expression, split andconversion of source and target images in different color spaces such as RGB, Grayand LMS to achieve its multi-source color transfer in l space. The comparisonand gray color channel selection tests of single and multi-source transfer prove thereasonability and effectiveness of the algorithm.
4. Intelligent optimized algorithm has different computational complexity withexponential growth and dependence on its multivariable coupling parameter settings,so the paper proposes Multi-Source Transfer Ant-Q and multi-source parametertransfer algorithm based on graph construction. The former analyzes the similarityratio between source and target tasks and determine each transfer samples by thisweight; the later constructs the model transfer graph of the source task includingknowledge (ACO operating parameters) to approximate the manifold space ofmultivariate parameters.
引文
[1] Meng J., Lin H., Li Y.. Knowledge transfer based on feature representation mapping for textclassification [J]. Expert Systems with Applications,2011,38(8):10562-10567.
[2] Deng J., Zhang Z., Marchi E., et al. Sparse autoencoder-based feature transfer learning forspeech emotion recognition [C]. In Proceedings of the2013Humaine AssociationConference on Affective Computing and Intelligent Interaction,2013:511-516.
[3] Shahriar M., Scott-Fleming I., Sari-Sarraf H., et al. A machine vision system to estimatecotton fiber maturity from longitudinal view using a transfer learning approach [J]. MachineVision and Applications,2013,24(8):1661-1683.
[4] Yoshida T., Ogino H.. Theoretical analysis and evaluation of topic graph based transferlearning [C]. In Proceedings of the9th International Conference on Active Media Technology,2013:106-115.
[5] Joo K. H., Park N. H., Choi J. T.. An adaptive teaching and learning system for efficientubiquitous learning [C]. In Proceedings of the8th International Conference on UbiquitousInformation Technologies and Applications,2014:659-666.
[6] Zheng D., Zhang C., Fei G., et al. Research on text categorization based on aweakly-supervised transfer learning method [C]. In Proceedings of the13th InternationalConference on Computational Linguistics and Intelligent,2012:144-156.
[7] Yang Q., Chen Y., Xue G., et al. Heterogeneous transfer learning for image clustering via thesocial web [C]. In Proceedings of the47th Annual Meeting of the Association forComputational Linguistics and4th International Joint Conference on Natural LanguageProcessing,2009:1-9.
[8] Pan W., Xiang E. W., Yang Q.. Transfer learning in collaborative filtering with uncertainratings [C]. In Proceedings of the26th AAAI Conference on Artificial Intelligence and the24th Innovative Applications of Artificial Intelligence Conference,2012:662-668.
[9] Zhang K., Wang Q., Lan L., et al. Sparse semi-supervised learning on low-rank kernel [J].Neurocomputing,2014,129:265-272.
[10] Xie C., Tan J., Chen P., et al. Collaborative object tracking model with local sparserepresentation [J]. Journal of Visual Communication and Image Representation,2014,25(2):423-434.
[11] Sun S., Hardoon D. R.. Active learning with extremely sparse labeled examples [J].Neurocomputing,2010,73(16-18):2980-2988.
[12] Yang Q. Transfer Learning beyond Text Classification [C]. In Proceedings of the1st AsianConference on Machine Learning,2009:10-22.
[13] Turing A. M.. Computing machine and intelligence [J]. Mind,1950,59(236):433-460.
[14] Sloman A.. The computer revolution in philosophy [M]. New York: Harvester Press andHumanities Press,1978.
[15] Thagard P. Computational philosophy of science [M]. Cambridge: MIT Press,1988.
[16] Minsky M. Society of mind [M]. Cambrdige: MIT Press,1988.
[17] Davis J., Domingos P.. Deep transfer via second-order Markov logic [C]. In Proceedings ofthe AAAI-2008Workshop on Transfer Learning for Complex Tasks,2008.
[18] Silver D. L.. Machine lifelong learning: challenges and benefits for artificial generalintelligence [C]. In Proceedings of4th International Conference on Artificial GeneralIntelligence,2011:370-375
[19] Pan S. J., Yang Q.. A survey on transfer learning [J]. IEEE Transactions on Knowledge andData Engineering,2010,22(10):1345-1359.
[20] Raina R., Battle A., Lee H., et al. Self-taught learning: transfer learning from unlabeled data[C]. In Proceedings of24th International Conference on Machine Learning,2007:759-766.
[21] Doan A., Madhavan J., Domingos P., et al. Ontology matching: A machine learning approach[C]. In Staab, S., and Studer, R., eds., Handbook on Ontologies in Information Systems.Springer-Velag,2004:397–416.
[22] Drineas P., Mahoney M. W.. On the Nystrffom method for approximating a Gram matrix forimproved kernel-based learning [J]. Journal of Machine Learning Research,2005,6:2153–2175.
[23] Duffy D. E., Santner T. J.. On the small sample properties of normrestricted maximumlikelihood estimators for logistic regression models [J]. Communications in Statistics: Theoryand Methods,1989,18:959–980.
[24] Falkenhainer B., Forbus K. D., Gentner D.. The structure-mapping engine: Algorithm andexamples [J]. Artificial Intelligence,1989,41(1):1–63.
[25] Eaton E., Desjardins M.. Set-based boosting for instance-level transfer [C]. In Proceedings of2009IEEE International Conference on Data Mining Workshops,2009:422-428.
[26] Fowlkes C., Belongie S., Chung F., et al. Spectral grouping using the Nystr om method [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(2):214–225.
[27] Freund Y., Schapire R. E.. A decision-theoretic generalization of online learning and anapplication to boosting [J]. Journal of Computer and System Sciences,1997,55(1):119–139.
[28] Gao J., Fan W., Jiang J., et al. Knowledge transfer via multiple model local structure mapping[C]. In Proceedings of the14th International Conference on Knowledge Discovery and DataMining,2008:283–291.
[29] Li D., Cao P., Guo Y., et al. Time weight update model based on the memory principle incollaborative filtering [J]. Journal of Computers,2013,8(11):2763-2767.
[30] Li Z., Wu X., Lu Z.. Fast orthogonal nonnegative matrix tri-factorization for simultaneousclustering [C]. In Proceedings of14th Pacific-Asia Conference on Knowledge Discovery andData Mining,2010:214-221.
[31] Reyzin L., Schapire R. E.. How boosting the margin can also boost classifier complexity [C].In Proceedings of the23rd International Conference on Machine Learning,2006:753–760.
[32] Dai W. Y., Xue G. R., Yang Y. Q.,Yu Y.. Co-clustering based classification for out-of-domaindocuments [C]. In Proceedings of the13th ACM SIGKDD international conference onKnowledge discovery and data mining,2007:210-219.
[33] Slimene A., Zagrouba E.. Kernel maximum mean discrepancy for region merging approach[C]. In Proceedings of15th International Conference on Computer Analysis of Images andPatterns,2013:475-482.
[34] Diu M., Gangeh M., Kamel M. S.. Unsupervised visual changepoint detection usingmaximum mean discrepancy. In Proceedings of10th International Conference on ImageAnalysis and Recognition,2013:336-345.
[35] Pan S. J., Yang Q.. A survey on transfer learning [J]. IEEE Transactions on Knowledge andData Engineering,2010,22(10):1345-1359.
[36] Shi X., Fan W., and Ren J.. Actively transfer domain knowledge [C]. In Proceedings of the19th European Conference on Machine Learning,2008:342–357.
[37] Stracuzzi D. J.. Scalable knowledge acquisition through cumulative learning and memoryorganization. Ph.D. Dissertation [D], University of Massachusetts,2006.
[38] Swarup S., Ray S. R.. Cross-domain knowledge transfer using structured representations [C].In Proceedings of the21st National Conference on Artificial Intelligence,2006:506–511.
[39] Szummer M., Jaakkola T.. Partially labeled classification with Markov random walks [C]. InAdvances in Neural Information Processing Systems,2002:945–952.
[40] Tanaka F., Yamamura M.. Multitask reinforcement learning on the distribution of MDPs [J].Transactions of the Institute of Electrical Engineers of Japan,2003,123(5):1004-1011.
[41] Evgeniou T., Pontil M.. Regularized multi–task learning [C]. In Proceedings of the10th ACMSIGKDD International Conference on Knowledge Discovery and Data Mining,2004:109-117.
[42] Taylor M. E., Whiteson S., Stone P.. Transfer via inter-task mappings in policy searchreinforcement learning [C]. In Proceedings of the Sixth International Joint Conference onAutonomous Agents and Multiagent Systems,2007:156–163.
[43] Thrun S., Mitchell T. M.. Learning one more thing [M]. Technical Report CMU-CS-94-184,Carnegie Mellon University, Pittsburgh, PA.1994.
[44] Thrun S., O’Sullivan J.. Clustering learning tasks and the selective crosstask transfer ofknowledge. Technical Report CMU-CS-95-209, Carnegie Mellon University, Pittsburgh, PA.1995.
[45] Thrun S., O’Sullivan J.. Discovering structure in multiple learning tasks: the TC algorithm
[C]. In Proceedings of the Thirteenth International Conference on Machine Learning,1996:489–497.
[46] Thrun S.. Explanation-based neural network learning: A Lifelong Learning Approach [M].Boston: Kluwer Academic Publishers,1996.
[47] Turk M., Pentland A.. Eigenfaces for recognition [J]. Cognitive Neuroscience,1991,3(1):71–86.
[48] Utgoff P. E., Stracuzzi D. J.. Many-layered learning [J].Neural Computation.2002,14:2497–2539.
[49] Lampert C. H., Kromer O.. Weakly-paired maximum covariance analysis for multimodaldimensionality reduction and transfer learning [C]. In Proceedings of11th EuropeanConference on Computer Vision,2010:566-579.
[50] Bickel S., Brückner M., Scheffer T.. Discriminative learning under covariate shift [J]. Journalof Machine Learning Research,2009,10(9):2137-2155.
[51] Zadrozny B.. Learning and evaluating classifiers under sample selection bias [C]. InProceedings of the twenty-first international conference on Machine learning,2004:114-119.
[52] Xiang E. W., Cao B., Hu D. H., et al. Bridging domains using world wide knowledge fortransfer learning [J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(6):770-783.
[53] Zhou D., Bousquet O., Lal T. N., et al. Learning with local and global consistency. InAdvances in Neural Information Processing Systems, MIT Press.2004,321–328.
[54] Zhou D., Huang J., Schffolkopf B.. Learning from labeled and unlabeled data on a directedgraph [C]. In Proceedings of the22nd International Conference on Machine Learning,2005:1036–1043.
[55] Yang J., Yan R., Alexander G. H.. Cross-domain video concept detection using adaptive svms[C]. In Proceedings of the15th international conference on Multimedia,2007:188-197.
[56] Yildirim H. M., Mukkai S. Krishnamoorthy. A random walk method for alleviating thesparsity problem in collaborative filtering [C]. In Proceedings of the2008ACM conferenceon Recommender systems,2008:131-138
[57] Yoo J., Choi S. J.. Probabilistic matrix tri-factorization [C]. In Proceedings of the2009IEEEInternational Conference on Acoustics, Speech and Signal Processing, ICASSP’09,2009:1553-1556.
[58] Yoo J., Choi S. J.. Weighted nonnegative matrix co-tri-factorization for collaborativeprediction [C]. In Proceedings of the1st Asian Conference on Machine Learning: Advancesin Machine Learning,2009:396-411.
[59] Zadrozny B.. Learning and evaluating classifiers under sample selection bias [C]. InProceedings of the twenty-first international conference on Machine learning,2004:114-119.
[60] Zhang L., Agarwal D., Chen B. C.. Generalizing matrix factorization through flexibleregression priors [C]. In Proceedings of the fifth ACM conference on Recommender systems,2011:13-20.
[61] Zhang Q., Qiu X. P., Huang X. J., Wu L.D.. Domain adaptation for conditional random fields[C]. In Proceedings of the4th Asia information retrieval conference on Information retrievaltechnology,2008:192-202
[62] Kumar A., Saha A., Daume Ⅲ H.. Co-regularization based semi-supervised domainadaptation [C]. In Preceedings of24th Annual Conference on Neural Information,2010:1-9.
[63] Blitzer J., Mcdonald R., Pereira F.. Domain adaptation with structural correspondencelearning [C]. In Proceedings of the2006Conference on Empirical Methods in NaturalLanguage Processing,2006:120-128.
[64] Ando R. K., Zhang T.. A framework for learning predictive structures from multiple tasks andunlabeled data [J]. Journal of Machine Learning Research,2005,6(12):1817-1853.
[65] Pan S. J., Kwok J. T., Yang Q.. Transfer learning via dimensionality reduction [C]. InProceedings of the23rd National Conference on Artificial Intelligence,2008:677-682.
[66] Karsten M. B., Arthur G., Malte J., et al. Integrating structured biological data by kernelmaximum mean discrepancy [C]. In Proceedings of the14th International Conference onIntelligent Systems for Molecular Biology,2006:49-57.
[67] Zhao S. W., Michelle X., Zhang X., et al. Who is doing what and when: Social map-basedrecommendation for content-centric social web sites [J]. ACM Transactions on IntelligentSystems and Technology (ACM TIST),3:5:1–5:23, October2011.
[68] Zheng Y., Xie X.. Learning travel recommendations from user-generated gps traces [J]. ACMTransactions on Intelligent Systems and Technology (ACM TIST),2:2:1–2:29, January2011.
[69] Zheng Y., Xie X... Learning travel recommendations from user-generated gps traces [J]. ACMTransactions on Intelligent Systems and Technology,2011,2(1):1–29.
[70] Arnold A., Nallapati R., Cohen W. W.. A comparative study of methods for transductivetransfer learning [C]. In Proceedings of the7th IEEE International Conference on DataMining Workshops,2007:77-82.
[71] Zhou T. C., Ma H., King I., Lyu M. R.. Tagrec: Leveraging tagging wisdom forrecommendation [C]. In Proceedings of the2009International Conference on ComputationalScience and Engineering-Volume04,2009:194-199.
[72] Dai W. Y., Xue G. R., Yang Y. Q.,Yu Y.. Co-clustering based classification for out-of-domaindocuments [C]. In Proceedings of the13th ACM SIGKDD international conference onKnowledge discovery and data mining,2007:210-219.
[73] Tanaka F., Yamamura M.. Multitask reinforcement learning on the distribution of MDPs [J].Transactions of the Institute of Electrical Engineers of Japan,2003,123(5):1004-1011.
[74] Lazaric A., Restelli M., Bonarini A.. Transfer of samples in batch reinforcement learning [C].In Proceedings of the25th International Conference on Machine Learning,2008:544-551.
[75] Sherstov A. A., Stone P.. Improving action selection in MDP's via knowledge transfer [C]. InProceedings of the20th National Conference on Artificial Intelligence and the17thInnovative Applications of Artificial Intelligence Conference,2005:1024-1029.
[76] Taylor M. E., Whiteson S., Stone P.. Transfer via inter-task mappings in policy searchreinforcement learning [C]. In Proceedings of the6th International Joint Conference onAutonomous Agents and Multiagent Systems,2007:156–163.
[77] Madden M. G., Howley T.. Transfer of experience between reinforcement learningenvironments with progressive difficulty [J], Artif. Intell. Rev.,2004,21(3):375-398.
[78] Fernandez F., Veloso M.. Probabilistic policy reuse in a reinforcement learning agent [C]. InProceedings of the5th International Conference on Autonomous Agents and Multi-AgentSystems,2006:720-727.
[79] Singh S. P.. Transfer of learning by composing solutions of elemental sequential tasks [J].Machine Learning,1992,8(3-4):323-339.
[80] Mehta N., Natarajan S., Tadepalli P., et al. Transfer in variable-reward hierarchicalreinforcement learning [J]. Mach. Learn.,2008,73(3):289-312.
[81] Dredze M., Crammer K.. Online methods for multi-domain learning and adaptation [C]. InProceedings of the Conference on Empirical Methods in Natural Language Processing,2008:689-697.
[82] Dredze M., Kulesza A., Crammer K.. Multi-domain learning by confidence-weightedparameter combination [J]. Machine Learning,2010,79:123–149.
[83] Duan L. X., Tsang I. W., Xu D., et al. Domain adaptation from multiple sources via auxiliaryclassifiers [C]. In Proceedings of the26th Annual International Conference on MachineLearning,2009:289-296.
[84] Asadi M., Hubber M.. Effective control knowledge transfer through learning skill andrepresentation hierarchies [C]. In Proceedings of the20th International Joint Conference onArtificial Intelligence,2007:2054-2059.
[85] Walsh T. J., Li J., Littman M. J.. Transferring state abstractions between MDPs [C]. InProceedings of ICML Workshop on Structural Knowledge Transfer for Machine Learning,2006.
[86] Taylor M. E., Stone P.. Transfer learning for reinforcement learning domains: a survey [J]. J.Mach. Learn. Res.,2009,10:1633-1685.
[87] Sholeh N., Lucian B., Robert B.. Efficient knowledge transfer in shaping reinforcementlearning [C]. In Proceedings of the18th IFAC World Congress,2011:8981-8986.
[88] Reinhard E., Adhikhmin M., Gooch B., et al. Color transfer between images [J]. IEEEComputer Graphics and Applications,2001,21(5):34-41.
[89] Zhang E., Zhang Y.. Color transfer algorithm based on mean-shift clustering [J]. Journal ofXi'an University of Technology,2009,25(1):105-109.
[90] Chang Y., Satio S., Nakajima M.. Color transfer between images based on basic colorcategory [C]. In Proceedings of IEICE Transactions on Information and Systems,2003:2780-2785.
[91] Tai Y. W., Jia J., Tang C. K.. Soft color segmentation and its applications [J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2007,29(9):1520-1537.
[92] Xiao X., Ma L.. Gradient-preserving color transfer [J]. Computer Graphics Forum,2009,28(7):1879-1886.
[93] Pitie F., Kokaram A. C., Dahyot R.. Automated colour grading using colour distributiontransfer [J]. Computer Vision and Image Understanding,2007,107(1-2):123-137.
[94] Chang Y., Saito S., Nakajima M.. Color transfer between images based on basic colorcategory [C]. In Proceedings of IEICE Transactions on Information and Systems,2003:2780-2785.
[95] Tai Y. W., Jia J., Tang C. K.. Local color transfer via probabilistic segmentation byexpectation-maximization [C]. In Proceedings of the IEEE Computer Society Conference onComputer Vision and Pattern Recognition,2005:747-754.
[96] Wen W., Fu D. M.. Colorization of infrared images based on DWT fusion and color transfer[C]. In Proceedings of the2007International Conference on Wavelet Analysis and Pattern,2007:432-436.
[97] Welsh T., Ashikhmin M., Mueller K.. Transferring color to greyscale images [C]. InProceedings of29th International Conference on Computer Graphics and InteractiveTechniques,2002:277-280.
[98] Xiao X., Ma L.. Color transfer in correlated color space [C]. In Proceedings of the ACMInternational Conference on Virtual Reality Continuum and its Applications,2006:305-309.
[99] Pitle F., Kokaram A. C., Dahyot R.. N-dimensional probability density function transfer andits application to color transfer [C]. In Proceedings of the10th IEEE International Conferenceon Computer Vision,2005:1434-1439.
[100] Vavilin A., Jo K. H.. Fast HDR image generation from multi-exposed multiple-view LDRimages [C]. In Proceedings of3rd European Workshop on Visual Information Processing,2011:105-110.
[101] Meng M., Liu L.. Sketching local color transfer [J]. Journal of Computer-Aided Design andComputer Graphics,2008,20(7):838-842.
[102] Guy I., Zwerdling N., Carmel D., et al. Personalized recommendation of social softwareitems based on social relations [C]. In Proceedings of the third ACM conference onRecommender systems,2009:53-60.
[103] Hannon J., Bennett M., Smyth B.. Recommending twitter users to follow using content andcollaborative filtering approaches [C]. In Proceedings of the fourth ACM conference onRecommender systems,2010:199-206
[104] Harshman R.. Foundations of the parafac procedure: Models and conditions for an“explanatory” multi-modal factor analysis [J]. UCLA Working Papers in Phonetics,1970,16:1–84.
[105] Freedman D., Kisilev P.. Object-to-object color transfer: optimal flows and SMSPtransformations [C]. In Proceedings of2010IEEE Conference on Computer Vision andPattern Recognition,2010:297-284.
[106] Hinrichs T. R., Forbus K. D.. Transfer learning through analogy in games [J]. AI Magazine,2011,32(1):70–83,
[107] Zaitar R. A., Hiyassat H.. Optimizing the ant colony optimization using standard geneticalgorithm [C]. In Proceedings of the23rd International Conference on Artificial Intelligenceand Applications,2005:130-134.
[108] Zhou Z. G.. An improved ant colony optimization supervised by PSO [J]. AdvancedMaterials Research,2010,108-111(1):1354-1359.
[109] Gambardella L. M., Dorigo M.. Ant-Q: a reinforcement learning approach to the travelingsalesman problem [C]. In Proceedings of12th International Conference on Machine Learning,1995:252-260.
[110] Cheng Y. H., Feng H. T., Wang X. S.. Actor-Critic learning based on adaptive importancesampling [J]. Chinese Journal of Electronics,2010,19(4):583-588.
[111] Rais H. M., Othman Z. A., Hamdan A. R.. Improved dynamic ant colony system (DACS) onsymmetric traveling salesman problem [C]. In Proceedings of International Conference onIntelligent and Advanced Systems,2008:43-48.
[112] Vien N. A., Viet N. H., Lee S. G., et al. Obstacle avoidance path planning for mobile robotbased on Ant-Q reinforcement learning algorithm [C]. In Proceedings of the4th InternationalSymposium on Neural Networks,2007:704-713.
[113] Machado L., Schirru R.. The Ant-Q algorithm applied to the nuclear reload problem [J].International Journal of Annals of Nuclear Energy,2002,29(12):1455-1470.
[114] Mariano C. E., Morelos E.. A multiple objective Ant-Q algorithm for the design of waterdistribution irrigation [C]. In Proceedings of the Genetic and Evolutionary ComputationConference,1999:894-901.
[115] Liu X. J., Ni Z. H.. Ant-Q algorithm based optimization approach for process planning [C].In Proceedings of the8th IEEE International Conference on Control and Automation,2010:620-623.
[116] Ceci M., Appice A., Barile N., et al. Transductive learning from relational data [C]. InProceedings of the5th International Conference on Machine Learning and Data Mining inPattern Recognition,2007:324-338.
[117] Dai W. Y., Yang Q., Xue G. R., et al. Boosting for transfer learning [C]. In Proceedings of the24th International Conference on Machine Learning,2007:193-200.
[118] Yao Y., Doretto G.. Boosting for transfer learning with multiple sources [C]. In Proceedingsof the IEEE Computer Society Conference on Computer Vision and Pattern Recognition,2010:1855-1862.
[119] Kocer B., Arslan A.. Genetic transfer learning [J]. Expert Systems with Applications,2010,37(10):6997-7002.
[120] Mihalkova L., Huynh T., Mooney R. J.. Mapping and revising markov logic networks fortransfer learning [C]. In Proceedings of the22nd AAAI Conference on Artificial Intelligenceand the19th Innovative Applications of Artificial Intelligence Conference. Vancouver,2007:608-614.
[121] Yu K., Chu W.. Gaussian process models for link analysis and transfer learning [C]. InProceedings of Annual Conference on Neural Information Processing Systems,2007:1-8.
[122] Lee J. W., Giraud C. C.. Transfer learning in decision trees [C]. In Proceedings ofInternational Joint Conference on Neural Networks,2007:726-731.
[123] Torrey L., Shavlik J., Walker T., et al. Relational macros for transfer in reinforcementlearning [C]. In Proceedings of the17th International Conference on Inductive LogicProgramming,2008:254-268.
[124] Turk M. A., Pentland A. P.. Face recognition using eigenfaces [C]. In Proceedings of IEEEConference on Computer Vision and Pattern Recognition,1991:586-591.
[125] Wu J., Zhou Z. H.. Face recognition with one training image per person [J]. PatternRecognit. Lett.,2002,23(14):1711-1719.
[126] Ko M., Barkana A.. A new solution to one sample problem in face recognition using FLDA[J]. Appl. Math. Comput.,2011,217(24):10368-10376.
[127] Fumera G., Roli F.. A theoretical and experimental analysis of linear combiners for multipleclassifer systems [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(6):942-956.
[128] Ibrahim R., Zin Z. M.. Study of automated face recognition system for office door accesscontrol application [C]. In Proceedings of IEEE Int. Conf. Commun,2011:132-136.
[129] Kumar V. K. N., Srinivasan B.. Enhancement of security and privacy in biometric passportinspection system using face, fingerprint, and rris recognition [J]. Int. J. Comput. Netw. Inf.Secur.,2012,4(8):55-64.
[130] Zhou M., Liang L., Sun J.. AAM based face tracking with temporal matching and facesegmentation [C]. In Proceedings of IEEE Comput. Soc. Conf. Comput. Vision PatternRecognit.,2010:701-708.
[131] Wu J., Zhou Z. H.. Face recognition with one training image per person [J]. PatternRecognit. Lett.,2002,23(14):1711-1719.
[132] Lee J. W., Giraud C. C.. Transfer learning in decision trees [C]. In Proceedings ofInternational Joint Conference on Neural Networks,2007:726-731.
[133] Lee S. W., Jung H. C., Hwang B.. Authenticating corrupted photo images based on noiseparameter estimation[J]. Pattern Recognit.,2006,39(5):910-920.
[134] Fawcett T.. An introduction to ROC analysis [J]. Pattern Recognition Letters,2006,27(8):861-874.
[135] Argyriou A., Evgeniou T., Pontil M.. Multi-task feature learning [C]. In Proceedings ofAdv. Neural Inf. Process. Syst.,2007:41-48.
[136] Mohammadzade H., Hatzinakos D.. Projection into expression subspaces for facerecognition from single sample per person [J]. IEEE Trans. Affect. Comput.,2013,4(1):69-82.
[137] Lee S. I., Chatalbashev V., Vickrey D., et al. Learning a meta-level prior for featurerelevance from multiple related tasks [C]. In Proceedings of International Conference onMachine Learning,2007:489-496.
[138] Jebara T.. Multi-task feature and kernel selection for svms [C]. In Proceedings ofInternational Conference on Machine Learning,2004:433-440.
[139] Ruckert U., Kramer S.. Kernel-based inductive transfer [C]. In Proceedings of EuropeanConference on Machine Learning and Knowledge Discovery,2008:220-233.
[140] Belhumeur P. N., Hespanha J. P., Kriegman D. J.. Eigenfaces vs. Fisherfaces: recognitionusing class specific linear projection [J]. IEEE Trans. Pattern Anal. Mach. Intell.,1997,19(7):711-720.
[141] He X., Yan S., Hu Y., et al. Face recognition using Laplacianfaces [J]. IEEE Trans. PatternAnal. Mach. Intell.,2005,27(3):328-340.
[142] Tang D., Zhu N., Yu F., et al. A novel sparse representation method based on virtual samplesfor face recognition [J]. Neural Comput. Appl.,2012,24(3):1-7.
[143] Yan L., Pan J..Face recognition with one sample per person based on contourlet and nearestfeature line [C]. In Proceedings of Int. Image Processing, Conf. Comput.&Patten Recogn.,2011:484-487.
[144] Li S., Jing X., Zhang D., et al. A novel kernel discriminant feature extraction frameworkbased on mapped virtual samples for face recognition [C]. In Proceedings of Int. Conf. ImageProcessing,2011:3005-3008.
[145] Liu C., Wechsler H.. Gabor feature based classification using the enhanced fisher lineardiscriminant model for face recognition [J]. IEEE Trans. Image Process.,2002,11(4):467-476.
[146] Hu J., Deng W., Guo J.. Robust discriminant analysis of latent semantic feature for textcategorization [C]. In Proceedings of International Conference on Fuzzy System andKnowledge Disccovery,2006:400-409.
[147] Lades M., Vorbruggen J., Buhmann J., et al. Distortion invariant object recognition in thedynamic link architecture [J]. IEEE Trans. Comput.,1993,42(3):300-311.
[148] Liu C.. The Bayes decision rule induced similarity measures [J]. IEEE Trans. Pattern Anal.Mach. Intell,2007,29(6):1086-1090.
[149] Ramon J., Driessens K., Croonenborqhs T.. Transfer learning in reinforcement learningproblems through partial policy [C]. In Proceedings of Lecture Notes in Artificial Intelligence,2007:699-707.
[150] Moradi S. A., Moradi H., Asadpour M..1-NN based approach for skill level estimation [C].In Proceedings of2012International Conference on Interactive Mobile and Computer AidedLearning,2012:192-196.
[151] Mehta N., Natarajan S., Tadepalli P., et al. Transfer in variable-reward hierarchicalreinforcement learning [J]. Mach. Learn.,2008,73(3):289-312.
[152] Huang D. S., Du J. X.. A constructive hybrid structure optimization methodology for radialbasis probabilistic neural networks [J]. IEEE Trans. Neural Netw.,2008,19(12):2099-2115.
[153] Fernandez F., Veloso M.. Probabilistic policy reuse in a reinforcement learning agent [C]. InProceedings of the5th International Conference on Autonomous Agents and Multi-AgentSystems,2006:720-727.
[154] Fernandez F., Veloso M.. Policy reuse for transfer learning across tasks with different stateand action spaces [C]. In Proceedings of the Workshop on Structural Knowledge Transfer forMachine Learning,2006.
[155] Pickett M., G.Barto A.. Policy blocks: an algorithm for creating useful macro-actions inreinforcement learning [C]. In Proceedings of the9th International Conference on MachineLearning,2002:506-513.
[156] Dietterich T. G.. Hierarchical reinforcement learning with the MAXQ value functiondecomposition [J]. J. Artif. Intell. Res.,2000,13(2):227-303.
[157] Huang D. S.. The local minima-free condition of feedforward neural networks forouter-supervised learning [J]. IEEE Trans. Syst. Man Cybern. Part B: Cybernetics.,1998,28(3):477-480.
[158] Mahadevan S.. Proto-value functions: developmental reinforcement learning [C]. InProceedings of the22nd International Conference on Machine Learning,2005:553-560.
[159] Madden M. G., Howley T.. Transfer of experience between reinforcement learningenvironments with progressive difficulty [J]. Artif. Intell. Rev.,2004,21(3):375-398.
[160] Driessens K., Ramon K. J., Croonenborghs T.. Transfer learning for reinforcement learningthrough goal and policy parameterization [C]. In Proceedings of the Workshop on StructuralKnowledge Transfer for Machine Learning,2006:1-4.
[161] Jouffe L.. Fuzzy inference system learning by reinforcement methods [J]. IEEE Trans. Syst.Man Cybern, Part C: Applications and Reviews.,1998,28(3):338-355.
[162] Jiang Y. Z., Deng Z. H., Wang S. T.. Mamdani-Larsen type transfer learning fuzzy system[J]. Acta Automatica Sinica,2012,38(9):1393-1490.
[163] Reinhard E., Adhikhmin M., Gooch B., et al. Color transfer between images [J]. IEEEComputer Graphics and Applications,2001,21(5):34-41.
[164] Zhang M., Ren J.. Driving and image enhancement for CCD sensing image system [C]. InProceedings of the3rd IEEE International Conference on Computer Science and InformationTechnology,2010:216-221.
[165] Wang W., Xu Y.. Color transfer algorithm in medical images [C]. In Proceedings of theInternational Society for Optical Engineering,2007:1-5.
[166] Rouf M., Lau C., Heidrich W.. Gradient domain color restoration of clipped highlights [C].In Proceedings of the IEEE Computer Society Conference on Computer Vision and PatternRecognition Workshops,2012:7-14.
[167] Lissiner I., Urban P.. Upgrading color-difference formulas [J]. Journal of the Optical Societyof America A: Optics and Image Science, and Vision,2010,27(7):1620-1629.
[168] Zeng K., Zhang R. M., Lan X. D., et al. Color style transfer by constraint locally linearembedding [C]. In Proceedings of the18th IEEE International Conference on ImageProcessing,2011:1121-1124.
[169] Tai Y. W., Jia J., Tang C. K.. Local color transfer via probabilistic segmentation byexpectation-maximization [C]. In Proceedings of the IEEE Computer Society Conference onComputer Vision and Pattern Recognition,2005:747-754.
[170] Xiang Y., Zou B. J., Wang H., et al. Multi-source color transfer for natural images [C]. InProceedings of the IEEE International Conference on Image Processing,2008:469-472.
[171] Guo Y. J., Li H., Zhang W., et al. Multi-source color transfer based on multi-labeled decisiontree [C]. In Proceedings of the9th International Conference for Young Computer Scientists.Hunan,2008:820-825.
[172] Oksanen J., Lundén J., Koivunen V.. Reinforcement learning based sensing policyoptimization for energy efficient cognitive radio networks [J]. Neurocomputing,2012,80(2):102-110.
[173] Chan T. F., Vese L. A.. Active contours without edges [J]. IEEE Transactions on ImageProcessing,2001,10(2):266-277.
[174] Hasler D., Süsstrunk S.. Measuring colourfulness in natural images [C]. In Proceedings ofthe International Society for Optical Engineering,2003:87-95.
[175] Eric E., Marie D. J., Terran L.. Modeling transfer relationships between learning tasks forimproved inductive transfer [C]. In Proceedings of Lectures Notes in Artificial Intelligence,2008:317-332.
[176] Wang X. R., Wu T. J.. The Ant(λ) ant colony optimization algorithm based on eligibilitytrace [C]. In Proceedings of the IEEE International Conference on Systems, Man andCybernetics,2003:4065-4070.
[177] Lee S. G., Chung T. C.. A reinforcement learning algorithm using temporal diference errorin ant model [C]. In Proceedings of Lectures Notes in Computer Science,2005:217-224.
[178] Pan S. J., Yang Q.. A survey on transfer learning [J]. IEEE Transactions on Knowledge andData Engineering,2010,22(10):1345-1359.
[179] Wang H., Gao Y., Chen X. G.. Transfer of reinforcement learning: the state of the art [J].Acta Electronica Sinica,2008,36(12):39-43.
[180] Eric E., Marie D. J., Terran L.. Modeling transfer relationships between learning tasks forimproved inductive transfer [C]. In Proceedings of Lectures Notes in Artificial Intelligence,2008:317-332.
[2] Deng J., Zhang Z., Marchi E., et al. Sparse autoencoder-based feature transfer learning forspeech emotion recognition [C]. In Proceedings of the2013Humaine AssociationConference on Affective Computing and Intelligent Interaction,2013:511-516.
[3] Shahriar M., Scott-Fleming I., Sari-Sarraf H., et al. A machine vision system to estimatecotton fiber maturity from longitudinal view using a transfer learning approach [J]. MachineVision and Applications,2013,24(8):1661-1683.
[4] Yoshida T., Ogino H.. Theoretical analysis and evaluation of topic graph based transferlearning [C]. In Proceedings of the9th International Conference on Active Media Technology,2013:106-115.
[5] Joo K. H., Park N. H., Choi J. T.. An adaptive teaching and learning system for efficientubiquitous learning [C]. In Proceedings of the8th International Conference on UbiquitousInformation Technologies and Applications,2014:659-666.
[6] Zheng D., Zhang C., Fei G., et al. Research on text categorization based on aweakly-supervised transfer learning method [C]. In Proceedings of the13th InternationalConference on Computational Linguistics and Intelligent,2012:144-156.
[7] Yang Q., Chen Y., Xue G., et al. Heterogeneous transfer learning for image clustering via thesocial web [C]. In Proceedings of the47th Annual Meeting of the Association forComputational Linguistics and4th International Joint Conference on Natural LanguageProcessing,2009:1-9.
[8] Pan W., Xiang E. W., Yang Q.. Transfer learning in collaborative filtering with uncertainratings [C]. In Proceedings of the26th AAAI Conference on Artificial Intelligence and the24th Innovative Applications of Artificial Intelligence Conference,2012:662-668.
[9] Zhang K., Wang Q., Lan L., et al. Sparse semi-supervised learning on low-rank kernel [J].Neurocomputing,2014,129:265-272.
[10] Xie C., Tan J., Chen P., et al. Collaborative object tracking model with local sparserepresentation [J]. Journal of Visual Communication and Image Representation,2014,25(2):423-434.
[11] Sun S., Hardoon D. R.. Active learning with extremely sparse labeled examples [J].Neurocomputing,2010,73(16-18):2980-2988.
[12] Yang Q. Transfer Learning beyond Text Classification [C]. In Proceedings of the1st AsianConference on Machine Learning,2009:10-22.
[13] Turing A. M.. Computing machine and intelligence [J]. Mind,1950,59(236):433-460.
[14] Sloman A.. The computer revolution in philosophy [M]. New York: Harvester Press andHumanities Press,1978.
[15] Thagard P. Computational philosophy of science [M]. Cambridge: MIT Press,1988.
[16] Minsky M. Society of mind [M]. Cambrdige: MIT Press,1988.
[17] Davis J., Domingos P.. Deep transfer via second-order Markov logic [C]. In Proceedings ofthe AAAI-2008Workshop on Transfer Learning for Complex Tasks,2008.
[18] Silver D. L.. Machine lifelong learning: challenges and benefits for artificial generalintelligence [C]. In Proceedings of4th International Conference on Artificial GeneralIntelligence,2011:370-375
[19] Pan S. J., Yang Q.. A survey on transfer learning [J]. IEEE Transactions on Knowledge andData Engineering,2010,22(10):1345-1359.
[20] Raina R., Battle A., Lee H., et al. Self-taught learning: transfer learning from unlabeled data[C]. In Proceedings of24th International Conference on Machine Learning,2007:759-766.
[21] Doan A., Madhavan J., Domingos P., et al. Ontology matching: A machine learning approach[C]. In Staab, S., and Studer, R., eds., Handbook on Ontologies in Information Systems.Springer-Velag,2004:397–416.
[22] Drineas P., Mahoney M. W.. On the Nystrffom method for approximating a Gram matrix forimproved kernel-based learning [J]. Journal of Machine Learning Research,2005,6:2153–2175.
[23] Duffy D. E., Santner T. J.. On the small sample properties of normrestricted maximumlikelihood estimators for logistic regression models [J]. Communications in Statistics: Theoryand Methods,1989,18:959–980.
[24] Falkenhainer B., Forbus K. D., Gentner D.. The structure-mapping engine: Algorithm andexamples [J]. Artificial Intelligence,1989,41(1):1–63.
[25] Eaton E., Desjardins M.. Set-based boosting for instance-level transfer [C]. In Proceedings of2009IEEE International Conference on Data Mining Workshops,2009:422-428.
[26] Fowlkes C., Belongie S., Chung F., et al. Spectral grouping using the Nystr om method [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2004,26(2):214–225.
[27] Freund Y., Schapire R. E.. A decision-theoretic generalization of online learning and anapplication to boosting [J]. Journal of Computer and System Sciences,1997,55(1):119–139.
[28] Gao J., Fan W., Jiang J., et al. Knowledge transfer via multiple model local structure mapping[C]. In Proceedings of the14th International Conference on Knowledge Discovery and DataMining,2008:283–291.
[29] Li D., Cao P., Guo Y., et al. Time weight update model based on the memory principle incollaborative filtering [J]. Journal of Computers,2013,8(11):2763-2767.
[30] Li Z., Wu X., Lu Z.. Fast orthogonal nonnegative matrix tri-factorization for simultaneousclustering [C]. In Proceedings of14th Pacific-Asia Conference on Knowledge Discovery andData Mining,2010:214-221.
[31] Reyzin L., Schapire R. E.. How boosting the margin can also boost classifier complexity [C].In Proceedings of the23rd International Conference on Machine Learning,2006:753–760.
[32] Dai W. Y., Xue G. R., Yang Y. Q.,Yu Y.. Co-clustering based classification for out-of-domaindocuments [C]. In Proceedings of the13th ACM SIGKDD international conference onKnowledge discovery and data mining,2007:210-219.
[33] Slimene A., Zagrouba E.. Kernel maximum mean discrepancy for region merging approach[C]. In Proceedings of15th International Conference on Computer Analysis of Images andPatterns,2013:475-482.
[34] Diu M., Gangeh M., Kamel M. S.. Unsupervised visual changepoint detection usingmaximum mean discrepancy. In Proceedings of10th International Conference on ImageAnalysis and Recognition,2013:336-345.
[35] Pan S. J., Yang Q.. A survey on transfer learning [J]. IEEE Transactions on Knowledge andData Engineering,2010,22(10):1345-1359.
[36] Shi X., Fan W., and Ren J.. Actively transfer domain knowledge [C]. In Proceedings of the19th European Conference on Machine Learning,2008:342–357.
[37] Stracuzzi D. J.. Scalable knowledge acquisition through cumulative learning and memoryorganization. Ph.D. Dissertation [D], University of Massachusetts,2006.
[38] Swarup S., Ray S. R.. Cross-domain knowledge transfer using structured representations [C].In Proceedings of the21st National Conference on Artificial Intelligence,2006:506–511.
[39] Szummer M., Jaakkola T.. Partially labeled classification with Markov random walks [C]. InAdvances in Neural Information Processing Systems,2002:945–952.
[40] Tanaka F., Yamamura M.. Multitask reinforcement learning on the distribution of MDPs [J].Transactions of the Institute of Electrical Engineers of Japan,2003,123(5):1004-1011.
[41] Evgeniou T., Pontil M.. Regularized multi–task learning [C]. In Proceedings of the10th ACMSIGKDD International Conference on Knowledge Discovery and Data Mining,2004:109-117.
[42] Taylor M. E., Whiteson S., Stone P.. Transfer via inter-task mappings in policy searchreinforcement learning [C]. In Proceedings of the Sixth International Joint Conference onAutonomous Agents and Multiagent Systems,2007:156–163.
[43] Thrun S., Mitchell T. M.. Learning one more thing [M]. Technical Report CMU-CS-94-184,Carnegie Mellon University, Pittsburgh, PA.1994.
[44] Thrun S., O’Sullivan J.. Clustering learning tasks and the selective crosstask transfer ofknowledge. Technical Report CMU-CS-95-209, Carnegie Mellon University, Pittsburgh, PA.1995.
[45] Thrun S., O’Sullivan J.. Discovering structure in multiple learning tasks: the TC algorithm
[C]. In Proceedings of the Thirteenth International Conference on Machine Learning,1996:489–497.
[46] Thrun S.. Explanation-based neural network learning: A Lifelong Learning Approach [M].Boston: Kluwer Academic Publishers,1996.
[47] Turk M., Pentland A.. Eigenfaces for recognition [J]. Cognitive Neuroscience,1991,3(1):71–86.
[48] Utgoff P. E., Stracuzzi D. J.. Many-layered learning [J].Neural Computation.2002,14:2497–2539.
[49] Lampert C. H., Kromer O.. Weakly-paired maximum covariance analysis for multimodaldimensionality reduction and transfer learning [C]. In Proceedings of11th EuropeanConference on Computer Vision,2010:566-579.
[50] Bickel S., Brückner M., Scheffer T.. Discriminative learning under covariate shift [J]. Journalof Machine Learning Research,2009,10(9):2137-2155.
[51] Zadrozny B.. Learning and evaluating classifiers under sample selection bias [C]. InProceedings of the twenty-first international conference on Machine learning,2004:114-119.
[52] Xiang E. W., Cao B., Hu D. H., et al. Bridging domains using world wide knowledge fortransfer learning [J]. IEEE Transactions on Knowledge and Data Engineering,2010,22(6):770-783.
[53] Zhou D., Bousquet O., Lal T. N., et al. Learning with local and global consistency. InAdvances in Neural Information Processing Systems, MIT Press.2004,321–328.
[54] Zhou D., Huang J., Schffolkopf B.. Learning from labeled and unlabeled data on a directedgraph [C]. In Proceedings of the22nd International Conference on Machine Learning,2005:1036–1043.
[55] Yang J., Yan R., Alexander G. H.. Cross-domain video concept detection using adaptive svms[C]. In Proceedings of the15th international conference on Multimedia,2007:188-197.
[56] Yildirim H. M., Mukkai S. Krishnamoorthy. A random walk method for alleviating thesparsity problem in collaborative filtering [C]. In Proceedings of the2008ACM conferenceon Recommender systems,2008:131-138
[57] Yoo J., Choi S. J.. Probabilistic matrix tri-factorization [C]. In Proceedings of the2009IEEEInternational Conference on Acoustics, Speech and Signal Processing, ICASSP’09,2009:1553-1556.
[58] Yoo J., Choi S. J.. Weighted nonnegative matrix co-tri-factorization for collaborativeprediction [C]. In Proceedings of the1st Asian Conference on Machine Learning: Advancesin Machine Learning,2009:396-411.
[59] Zadrozny B.. Learning and evaluating classifiers under sample selection bias [C]. InProceedings of the twenty-first international conference on Machine learning,2004:114-119.
[60] Zhang L., Agarwal D., Chen B. C.. Generalizing matrix factorization through flexibleregression priors [C]. In Proceedings of the fifth ACM conference on Recommender systems,2011:13-20.
[61] Zhang Q., Qiu X. P., Huang X. J., Wu L.D.. Domain adaptation for conditional random fields[C]. In Proceedings of the4th Asia information retrieval conference on Information retrievaltechnology,2008:192-202
[62] Kumar A., Saha A., Daume Ⅲ H.. Co-regularization based semi-supervised domainadaptation [C]. In Preceedings of24th Annual Conference on Neural Information,2010:1-9.
[63] Blitzer J., Mcdonald R., Pereira F.. Domain adaptation with structural correspondencelearning [C]. In Proceedings of the2006Conference on Empirical Methods in NaturalLanguage Processing,2006:120-128.
[64] Ando R. K., Zhang T.. A framework for learning predictive structures from multiple tasks andunlabeled data [J]. Journal of Machine Learning Research,2005,6(12):1817-1853.
[65] Pan S. J., Kwok J. T., Yang Q.. Transfer learning via dimensionality reduction [C]. InProceedings of the23rd National Conference on Artificial Intelligence,2008:677-682.
[66] Karsten M. B., Arthur G., Malte J., et al. Integrating structured biological data by kernelmaximum mean discrepancy [C]. In Proceedings of the14th International Conference onIntelligent Systems for Molecular Biology,2006:49-57.
[67] Zhao S. W., Michelle X., Zhang X., et al. Who is doing what and when: Social map-basedrecommendation for content-centric social web sites [J]. ACM Transactions on IntelligentSystems and Technology (ACM TIST),3:5:1–5:23, October2011.
[68] Zheng Y., Xie X.. Learning travel recommendations from user-generated gps traces [J]. ACMTransactions on Intelligent Systems and Technology (ACM TIST),2:2:1–2:29, January2011.
[69] Zheng Y., Xie X... Learning travel recommendations from user-generated gps traces [J]. ACMTransactions on Intelligent Systems and Technology,2011,2(1):1–29.
[70] Arnold A., Nallapati R., Cohen W. W.. A comparative study of methods for transductivetransfer learning [C]. In Proceedings of the7th IEEE International Conference on DataMining Workshops,2007:77-82.
[71] Zhou T. C., Ma H., King I., Lyu M. R.. Tagrec: Leveraging tagging wisdom forrecommendation [C]. In Proceedings of the2009International Conference on ComputationalScience and Engineering-Volume04,2009:194-199.
[72] Dai W. Y., Xue G. R., Yang Y. Q.,Yu Y.. Co-clustering based classification for out-of-domaindocuments [C]. In Proceedings of the13th ACM SIGKDD international conference onKnowledge discovery and data mining,2007:210-219.
[73] Tanaka F., Yamamura M.. Multitask reinforcement learning on the distribution of MDPs [J].Transactions of the Institute of Electrical Engineers of Japan,2003,123(5):1004-1011.
[74] Lazaric A., Restelli M., Bonarini A.. Transfer of samples in batch reinforcement learning [C].In Proceedings of the25th International Conference on Machine Learning,2008:544-551.
[75] Sherstov A. A., Stone P.. Improving action selection in MDP's via knowledge transfer [C]. InProceedings of the20th National Conference on Artificial Intelligence and the17thInnovative Applications of Artificial Intelligence Conference,2005:1024-1029.
[76] Taylor M. E., Whiteson S., Stone P.. Transfer via inter-task mappings in policy searchreinforcement learning [C]. In Proceedings of the6th International Joint Conference onAutonomous Agents and Multiagent Systems,2007:156–163.
[77] Madden M. G., Howley T.. Transfer of experience between reinforcement learningenvironments with progressive difficulty [J], Artif. Intell. Rev.,2004,21(3):375-398.
[78] Fernandez F., Veloso M.. Probabilistic policy reuse in a reinforcement learning agent [C]. InProceedings of the5th International Conference on Autonomous Agents and Multi-AgentSystems,2006:720-727.
[79] Singh S. P.. Transfer of learning by composing solutions of elemental sequential tasks [J].Machine Learning,1992,8(3-4):323-339.
[80] Mehta N., Natarajan S., Tadepalli P., et al. Transfer in variable-reward hierarchicalreinforcement learning [J]. Mach. Learn.,2008,73(3):289-312.
[81] Dredze M., Crammer K.. Online methods for multi-domain learning and adaptation [C]. InProceedings of the Conference on Empirical Methods in Natural Language Processing,2008:689-697.
[82] Dredze M., Kulesza A., Crammer K.. Multi-domain learning by confidence-weightedparameter combination [J]. Machine Learning,2010,79:123–149.
[83] Duan L. X., Tsang I. W., Xu D., et al. Domain adaptation from multiple sources via auxiliaryclassifiers [C]. In Proceedings of the26th Annual International Conference on MachineLearning,2009:289-296.
[84] Asadi M., Hubber M.. Effective control knowledge transfer through learning skill andrepresentation hierarchies [C]. In Proceedings of the20th International Joint Conference onArtificial Intelligence,2007:2054-2059.
[85] Walsh T. J., Li J., Littman M. J.. Transferring state abstractions between MDPs [C]. InProceedings of ICML Workshop on Structural Knowledge Transfer for Machine Learning,2006.
[86] Taylor M. E., Stone P.. Transfer learning for reinforcement learning domains: a survey [J]. J.Mach. Learn. Res.,2009,10:1633-1685.
[87] Sholeh N., Lucian B., Robert B.. Efficient knowledge transfer in shaping reinforcementlearning [C]. In Proceedings of the18th IFAC World Congress,2011:8981-8986.
[88] Reinhard E., Adhikhmin M., Gooch B., et al. Color transfer between images [J]. IEEEComputer Graphics and Applications,2001,21(5):34-41.
[89] Zhang E., Zhang Y.. Color transfer algorithm based on mean-shift clustering [J]. Journal ofXi'an University of Technology,2009,25(1):105-109.
[90] Chang Y., Satio S., Nakajima M.. Color transfer between images based on basic colorcategory [C]. In Proceedings of IEICE Transactions on Information and Systems,2003:2780-2785.
[91] Tai Y. W., Jia J., Tang C. K.. Soft color segmentation and its applications [J]. IEEETransactions on Pattern Analysis and Machine Intelligence,2007,29(9):1520-1537.
[92] Xiao X., Ma L.. Gradient-preserving color transfer [J]. Computer Graphics Forum,2009,28(7):1879-1886.
[93] Pitie F., Kokaram A. C., Dahyot R.. Automated colour grading using colour distributiontransfer [J]. Computer Vision and Image Understanding,2007,107(1-2):123-137.
[94] Chang Y., Saito S., Nakajima M.. Color transfer between images based on basic colorcategory [C]. In Proceedings of IEICE Transactions on Information and Systems,2003:2780-2785.
[95] Tai Y. W., Jia J., Tang C. K.. Local color transfer via probabilistic segmentation byexpectation-maximization [C]. In Proceedings of the IEEE Computer Society Conference onComputer Vision and Pattern Recognition,2005:747-754.
[96] Wen W., Fu D. M.. Colorization of infrared images based on DWT fusion and color transfer[C]. In Proceedings of the2007International Conference on Wavelet Analysis and Pattern,2007:432-436.
[97] Welsh T., Ashikhmin M., Mueller K.. Transferring color to greyscale images [C]. InProceedings of29th International Conference on Computer Graphics and InteractiveTechniques,2002:277-280.
[98] Xiao X., Ma L.. Color transfer in correlated color space [C]. In Proceedings of the ACMInternational Conference on Virtual Reality Continuum and its Applications,2006:305-309.
[99] Pitle F., Kokaram A. C., Dahyot R.. N-dimensional probability density function transfer andits application to color transfer [C]. In Proceedings of the10th IEEE International Conferenceon Computer Vision,2005:1434-1439.
[100] Vavilin A., Jo K. H.. Fast HDR image generation from multi-exposed multiple-view LDRimages [C]. In Proceedings of3rd European Workshop on Visual Information Processing,2011:105-110.
[101] Meng M., Liu L.. Sketching local color transfer [J]. Journal of Computer-Aided Design andComputer Graphics,2008,20(7):838-842.
[102] Guy I., Zwerdling N., Carmel D., et al. Personalized recommendation of social softwareitems based on social relations [C]. In Proceedings of the third ACM conference onRecommender systems,2009:53-60.
[103] Hannon J., Bennett M., Smyth B.. Recommending twitter users to follow using content andcollaborative filtering approaches [C]. In Proceedings of the fourth ACM conference onRecommender systems,2010:199-206
[104] Harshman R.. Foundations of the parafac procedure: Models and conditions for an“explanatory” multi-modal factor analysis [J]. UCLA Working Papers in Phonetics,1970,16:1–84.
[105] Freedman D., Kisilev P.. Object-to-object color transfer: optimal flows and SMSPtransformations [C]. In Proceedings of2010IEEE Conference on Computer Vision andPattern Recognition,2010:297-284.
[106] Hinrichs T. R., Forbus K. D.. Transfer learning through analogy in games [J]. AI Magazine,2011,32(1):70–83,
[107] Zaitar R. A., Hiyassat H.. Optimizing the ant colony optimization using standard geneticalgorithm [C]. In Proceedings of the23rd International Conference on Artificial Intelligenceand Applications,2005:130-134.
[108] Zhou Z. G.. An improved ant colony optimization supervised by PSO [J]. AdvancedMaterials Research,2010,108-111(1):1354-1359.
[109] Gambardella L. M., Dorigo M.. Ant-Q: a reinforcement learning approach to the travelingsalesman problem [C]. In Proceedings of12th International Conference on Machine Learning,1995:252-260.
[110] Cheng Y. H., Feng H. T., Wang X. S.. Actor-Critic learning based on adaptive importancesampling [J]. Chinese Journal of Electronics,2010,19(4):583-588.
[111] Rais H. M., Othman Z. A., Hamdan A. R.. Improved dynamic ant colony system (DACS) onsymmetric traveling salesman problem [C]. In Proceedings of International Conference onIntelligent and Advanced Systems,2008:43-48.
[112] Vien N. A., Viet N. H., Lee S. G., et al. Obstacle avoidance path planning for mobile robotbased on Ant-Q reinforcement learning algorithm [C]. In Proceedings of the4th InternationalSymposium on Neural Networks,2007:704-713.
[113] Machado L., Schirru R.. The Ant-Q algorithm applied to the nuclear reload problem [J].International Journal of Annals of Nuclear Energy,2002,29(12):1455-1470.
[114] Mariano C. E., Morelos E.. A multiple objective Ant-Q algorithm for the design of waterdistribution irrigation [C]. In Proceedings of the Genetic and Evolutionary ComputationConference,1999:894-901.
[115] Liu X. J., Ni Z. H.. Ant-Q algorithm based optimization approach for process planning [C].In Proceedings of the8th IEEE International Conference on Control and Automation,2010:620-623.
[116] Ceci M., Appice A., Barile N., et al. Transductive learning from relational data [C]. InProceedings of the5th International Conference on Machine Learning and Data Mining inPattern Recognition,2007:324-338.
[117] Dai W. Y., Yang Q., Xue G. R., et al. Boosting for transfer learning [C]. In Proceedings of the24th International Conference on Machine Learning,2007:193-200.
[118] Yao Y., Doretto G.. Boosting for transfer learning with multiple sources [C]. In Proceedingsof the IEEE Computer Society Conference on Computer Vision and Pattern Recognition,2010:1855-1862.
[119] Kocer B., Arslan A.. Genetic transfer learning [J]. Expert Systems with Applications,2010,37(10):6997-7002.
[120] Mihalkova L., Huynh T., Mooney R. J.. Mapping and revising markov logic networks fortransfer learning [C]. In Proceedings of the22nd AAAI Conference on Artificial Intelligenceand the19th Innovative Applications of Artificial Intelligence Conference. Vancouver,2007:608-614.
[121] Yu K., Chu W.. Gaussian process models for link analysis and transfer learning [C]. InProceedings of Annual Conference on Neural Information Processing Systems,2007:1-8.
[122] Lee J. W., Giraud C. C.. Transfer learning in decision trees [C]. In Proceedings ofInternational Joint Conference on Neural Networks,2007:726-731.
[123] Torrey L., Shavlik J., Walker T., et al. Relational macros for transfer in reinforcementlearning [C]. In Proceedings of the17th International Conference on Inductive LogicProgramming,2008:254-268.
[124] Turk M. A., Pentland A. P.. Face recognition using eigenfaces [C]. In Proceedings of IEEEConference on Computer Vision and Pattern Recognition,1991:586-591.
[125] Wu J., Zhou Z. H.. Face recognition with one training image per person [J]. PatternRecognit. Lett.,2002,23(14):1711-1719.
[126] Ko M., Barkana A.. A new solution to one sample problem in face recognition using FLDA[J]. Appl. Math. Comput.,2011,217(24):10368-10376.
[127] Fumera G., Roli F.. A theoretical and experimental analysis of linear combiners for multipleclassifer systems [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(6):942-956.
[128] Ibrahim R., Zin Z. M.. Study of automated face recognition system for office door accesscontrol application [C]. In Proceedings of IEEE Int. Conf. Commun,2011:132-136.
[129] Kumar V. K. N., Srinivasan B.. Enhancement of security and privacy in biometric passportinspection system using face, fingerprint, and rris recognition [J]. Int. J. Comput. Netw. Inf.Secur.,2012,4(8):55-64.
[130] Zhou M., Liang L., Sun J.. AAM based face tracking with temporal matching and facesegmentation [C]. In Proceedings of IEEE Comput. Soc. Conf. Comput. Vision PatternRecognit.,2010:701-708.
[131] Wu J., Zhou Z. H.. Face recognition with one training image per person [J]. PatternRecognit. Lett.,2002,23(14):1711-1719.
[132] Lee J. W., Giraud C. C.. Transfer learning in decision trees [C]. In Proceedings ofInternational Joint Conference on Neural Networks,2007:726-731.
[133] Lee S. W., Jung H. C., Hwang B.. Authenticating corrupted photo images based on noiseparameter estimation[J]. Pattern Recognit.,2006,39(5):910-920.
[134] Fawcett T.. An introduction to ROC analysis [J]. Pattern Recognition Letters,2006,27(8):861-874.
[135] Argyriou A., Evgeniou T., Pontil M.. Multi-task feature learning [C]. In Proceedings ofAdv. Neural Inf. Process. Syst.,2007:41-48.
[136] Mohammadzade H., Hatzinakos D.. Projection into expression subspaces for facerecognition from single sample per person [J]. IEEE Trans. Affect. Comput.,2013,4(1):69-82.
[137] Lee S. I., Chatalbashev V., Vickrey D., et al. Learning a meta-level prior for featurerelevance from multiple related tasks [C]. In Proceedings of International Conference onMachine Learning,2007:489-496.
[138] Jebara T.. Multi-task feature and kernel selection for svms [C]. In Proceedings ofInternational Conference on Machine Learning,2004:433-440.
[139] Ruckert U., Kramer S.. Kernel-based inductive transfer [C]. In Proceedings of EuropeanConference on Machine Learning and Knowledge Discovery,2008:220-233.
[140] Belhumeur P. N., Hespanha J. P., Kriegman D. J.. Eigenfaces vs. Fisherfaces: recognitionusing class specific linear projection [J]. IEEE Trans. Pattern Anal. Mach. Intell.,1997,19(7):711-720.
[141] He X., Yan S., Hu Y., et al. Face recognition using Laplacianfaces [J]. IEEE Trans. PatternAnal. Mach. Intell.,2005,27(3):328-340.
[142] Tang D., Zhu N., Yu F., et al. A novel sparse representation method based on virtual samplesfor face recognition [J]. Neural Comput. Appl.,2012,24(3):1-7.
[143] Yan L., Pan J..Face recognition with one sample per person based on contourlet and nearestfeature line [C]. In Proceedings of Int. Image Processing, Conf. Comput.&Patten Recogn.,2011:484-487.
[144] Li S., Jing X., Zhang D., et al. A novel kernel discriminant feature extraction frameworkbased on mapped virtual samples for face recognition [C]. In Proceedings of Int. Conf. ImageProcessing,2011:3005-3008.
[145] Liu C., Wechsler H.. Gabor feature based classification using the enhanced fisher lineardiscriminant model for face recognition [J]. IEEE Trans. Image Process.,2002,11(4):467-476.
[146] Hu J., Deng W., Guo J.. Robust discriminant analysis of latent semantic feature for textcategorization [C]. In Proceedings of International Conference on Fuzzy System andKnowledge Disccovery,2006:400-409.
[147] Lades M., Vorbruggen J., Buhmann J., et al. Distortion invariant object recognition in thedynamic link architecture [J]. IEEE Trans. Comput.,1993,42(3):300-311.
[148] Liu C.. The Bayes decision rule induced similarity measures [J]. IEEE Trans. Pattern Anal.Mach. Intell,2007,29(6):1086-1090.
[149] Ramon J., Driessens K., Croonenborqhs T.. Transfer learning in reinforcement learningproblems through partial policy [C]. In Proceedings of Lecture Notes in Artificial Intelligence,2007:699-707.
[150] Moradi S. A., Moradi H., Asadpour M..1-NN based approach for skill level estimation [C].In Proceedings of2012International Conference on Interactive Mobile and Computer AidedLearning,2012:192-196.
[151] Mehta N., Natarajan S., Tadepalli P., et al. Transfer in variable-reward hierarchicalreinforcement learning [J]. Mach. Learn.,2008,73(3):289-312.
[152] Huang D. S., Du J. X.. A constructive hybrid structure optimization methodology for radialbasis probabilistic neural networks [J]. IEEE Trans. Neural Netw.,2008,19(12):2099-2115.
[153] Fernandez F., Veloso M.. Probabilistic policy reuse in a reinforcement learning agent [C]. InProceedings of the5th International Conference on Autonomous Agents and Multi-AgentSystems,2006:720-727.
[154] Fernandez F., Veloso M.. Policy reuse for transfer learning across tasks with different stateand action spaces [C]. In Proceedings of the Workshop on Structural Knowledge Transfer forMachine Learning,2006.
[155] Pickett M., G.Barto A.. Policy blocks: an algorithm for creating useful macro-actions inreinforcement learning [C]. In Proceedings of the9th International Conference on MachineLearning,2002:506-513.
[156] Dietterich T. G.. Hierarchical reinforcement learning with the MAXQ value functiondecomposition [J]. J. Artif. Intell. Res.,2000,13(2):227-303.
[157] Huang D. S.. The local minima-free condition of feedforward neural networks forouter-supervised learning [J]. IEEE Trans. Syst. Man Cybern. Part B: Cybernetics.,1998,28(3):477-480.
[158] Mahadevan S.. Proto-value functions: developmental reinforcement learning [C]. InProceedings of the22nd International Conference on Machine Learning,2005:553-560.
[159] Madden M. G., Howley T.. Transfer of experience between reinforcement learningenvironments with progressive difficulty [J]. Artif. Intell. Rev.,2004,21(3):375-398.
[160] Driessens K., Ramon K. J., Croonenborghs T.. Transfer learning for reinforcement learningthrough goal and policy parameterization [C]. In Proceedings of the Workshop on StructuralKnowledge Transfer for Machine Learning,2006:1-4.
[161] Jouffe L.. Fuzzy inference system learning by reinforcement methods [J]. IEEE Trans. Syst.Man Cybern, Part C: Applications and Reviews.,1998,28(3):338-355.
[162] Jiang Y. Z., Deng Z. H., Wang S. T.. Mamdani-Larsen type transfer learning fuzzy system[J]. Acta Automatica Sinica,2012,38(9):1393-1490.
[163] Reinhard E., Adhikhmin M., Gooch B., et al. Color transfer between images [J]. IEEEComputer Graphics and Applications,2001,21(5):34-41.
[164] Zhang M., Ren J.. Driving and image enhancement for CCD sensing image system [C]. InProceedings of the3rd IEEE International Conference on Computer Science and InformationTechnology,2010:216-221.
[165] Wang W., Xu Y.. Color transfer algorithm in medical images [C]. In Proceedings of theInternational Society for Optical Engineering,2007:1-5.
[166] Rouf M., Lau C., Heidrich W.. Gradient domain color restoration of clipped highlights [C].In Proceedings of the IEEE Computer Society Conference on Computer Vision and PatternRecognition Workshops,2012:7-14.
[167] Lissiner I., Urban P.. Upgrading color-difference formulas [J]. Journal of the Optical Societyof America A: Optics and Image Science, and Vision,2010,27(7):1620-1629.
[168] Zeng K., Zhang R. M., Lan X. D., et al. Color style transfer by constraint locally linearembedding [C]. In Proceedings of the18th IEEE International Conference on ImageProcessing,2011:1121-1124.
[169] Tai Y. W., Jia J., Tang C. K.. Local color transfer via probabilistic segmentation byexpectation-maximization [C]. In Proceedings of the IEEE Computer Society Conference onComputer Vision and Pattern Recognition,2005:747-754.
[170] Xiang Y., Zou B. J., Wang H., et al. Multi-source color transfer for natural images [C]. InProceedings of the IEEE International Conference on Image Processing,2008:469-472.
[171] Guo Y. J., Li H., Zhang W., et al. Multi-source color transfer based on multi-labeled decisiontree [C]. In Proceedings of the9th International Conference for Young Computer Scientists.Hunan,2008:820-825.
[172] Oksanen J., Lundén J., Koivunen V.. Reinforcement learning based sensing policyoptimization for energy efficient cognitive radio networks [J]. Neurocomputing,2012,80(2):102-110.
[173] Chan T. F., Vese L. A.. Active contours without edges [J]. IEEE Transactions on ImageProcessing,2001,10(2):266-277.
[174] Hasler D., Süsstrunk S.. Measuring colourfulness in natural images [C]. In Proceedings ofthe International Society for Optical Engineering,2003:87-95.
[175] Eric E., Marie D. J., Terran L.. Modeling transfer relationships between learning tasks forimproved inductive transfer [C]. In Proceedings of Lectures Notes in Artificial Intelligence,2008:317-332.
[176] Wang X. R., Wu T. J.. The Ant(λ) ant colony optimization algorithm based on eligibilitytrace [C]. In Proceedings of the IEEE International Conference on Systems, Man andCybernetics,2003:4065-4070.
[177] Lee S. G., Chung T. C.. A reinforcement learning algorithm using temporal diference errorin ant model [C]. In Proceedings of Lectures Notes in Computer Science,2005:217-224.
[178] Pan S. J., Yang Q.. A survey on transfer learning [J]. IEEE Transactions on Knowledge andData Engineering,2010,22(10):1345-1359.
[179] Wang H., Gao Y., Chen X. G.. Transfer of reinforcement learning: the state of the art [J].Acta Electronica Sinica,2008,36(12):39-43.
[180] Eric E., Marie D. J., Terran L.. Modeling transfer relationships between learning tasks forimproved inductive transfer [C]. In Proceedings of Lectures Notes in Artificial Intelligence,2008:317-332.