Abstract To model associative relationships among multiple-source data in online way, an online associative memory model based on self-organizing decision tree is proposed with the consideration of the efficient computation performance and good noise robustness. In the proposed model, real multi-source data are firstly reduced into finite representatives for information enhancement. Then, data representatives are divided into different sub-domains based on decision tree algorithm. Finally, the associative relations among multi-source data are trained on different sub-domains. The learning stability of the proposed model is analyzed theoretically. The experimental results demonstrate the proposed model can gain good performance on online classification learning and hetero-associative modeling for noisy data.
About author:: XIE Zhenping(Corresponding author), born in 1979, Ph.D., associate professor. His research interests include machine learning and cognitive computing.SUN Tao, born in 1991, master student. His research interests include artificial intelligence and machine learning.
[1] MICHEL A N, FARRELL J A. Associative Memories via Artificial Neural Networks. IEEE Control Systems Magazine, 1990, 10(3): 6-17. [2] ORTIZ-RODRIGUEZ J M, DEL ROSARIO M B M, EDUARDO G, et al. Artificial Neural Networks Modeling Evolved Genetically, a New Approach Applied in Neutron Spectrometry and Dosimetry Research Areas // Proc of the Electronics, Robotics and Automotive Mechanics Conference. Washington, USA: IEEE, 2008: 387-392. [3] 刘智斌,曾晓勤,刘惠义,等.基于BP神经网络的双层启发式强化学习方法.计算机研究与发展, 2015, 52(3): 579-587. (LIU Z B, ZENG X Q, LIU H Y, et al. A Heuristic Two-Layer Reinforcement Learning Algorithm Based on BP Neural Networks. Journal of Computer Research & Development, 2015, 52(3): 579-587.) [4] FURAO S, HASEGAWA O. An Incremental Network for On-line Unsupervised Classification and Topology Learning. Neural Networks, 2006, 19(1): 90-106. [5] HOPFIELD J J. Neural Networks and Physical Systems with Emergent Collective Computational Abilities. Proceedings of the National Academy of Sciences of the United States of America, 1982, 79(8): 2554-2558. [6] WANG M, CHEN S. Enhanced FMAM Based on Empirical Kernel Map. IEEE Transactions on Neural Networks, 2005, 16(3): 557-564. [7] ANDRADE A O. Analysis of Fuzzy Morphology in Spore Counts of Mycorrhizal Fungi // Proc of the 5th World Conference on Soft Computing. Washington, USA: IEEE, 2015. DOI: 10.1109/NAFIPS-WConSC.2015.7284131. [8] LAKE B M, SALAKHUTDINOV R, TENENBAUM J B. Human-Level Concept Learning through Probabilistic Program Induction. Science, 2015, 350(6266): 1332-1338. [9] XIE Z P, SUN J, PALADE V, et al. Evolutionary Sampling: A Novel Way of Machine Learning within a Probabilistic Framework. Information Sciences, 2014, 299: 262-282. [10] 史春奇,施智平,刘 曦,等.基于自组织动态神经网络的图像分割.计算机研究与发展, 2009, 46(1): 23-30. (SHI C Q, SHI Z P, LIU X, et al. Image Segmentation Based on Self-organizing Dynamic Neural Network. Journal of Computer Research and Development, 2009, 46(1): 23-30.) [11] 王士同,谢振平,李涵雄.模糊推理的统计敏感性分析.智能系统学报, 2007, 2(2): 57-64. (WANG S T, XIE Z P, LI H X. Statistical Sensitivity Analysis of Fuzzy Reasoning. CAAI Transactions on Intelligent Systems, 2007, 2(2): 57-64.) [12] 谢振平,王 涛,刘 渊,等.一种飘动性分析的视频烟雾快速检测新算法.微电子学与计算机, 2011, 28(10): 209-214. (XIE Z P, WANG T, LIU Y, et al. A Fast Video Smoke Detection Algorithm by Analyzing Fluttering. Microelectronics & Computer, 2011, 28(10): 209-214.) [13] BULO S R, KONTSCHIEDER P. Neural Decision Forests for Semantic Image Labelling // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2014: 81-88. [14] KOHUTKA L, PISTEK P. Faster Synthesis of Combinational Logic Based on Multiplexer Trees and Binary Decision Diagrams // Proc of the 12th IEEE International Conference on Emerging Learning Technologies and Applications. Washington, USA: IEEE, 2014: 239-244. [15] 王建勇,周晓光,廖启征.一种基于中值-模糊技术的混合噪声滤波器.电子与信息学报, 2006, 28(5): 901-904. (WANG J Y, ZHOU X G, LIAO Q Z. Mixed Noise Filter Based on Median-Fuzzy Technology. Journal of Electronics & Information Technology, 2006, 28(5): 901-904. [16] SHEN F R, OUYANG Q B, KASSAI W, et al. A General Associative Memory Based on Self-organizing Incremental Neural Network. Neurocomputing, 2013, 104: 57-71. [17] 孙 桃,谢振平,王士同,等.容量约束的自组织增量联想记忆模型.计算机科学与探索, 2016(1): 130-141. (SUN T, XIE Z P, WANG S T, et al. Self-organizing Incremental Associative Memory Model under Capacity Constraint. Journal of Frontiers of Computer Science and Technology, 2016(1): 130-141. [18] 吴 蕾,张文生,王 珏.基于深度学习框架的隐藏主题变量图模型.计算机研究与发展, 2015, 52(1): 191-199. (WU L, ZHANG W S, WANG J. Hidden Topic Variable Graphical Model Based on Deep Learning Framework. Journal of Computer Research and Development, 2015, 52(1): 191-199. [19] BRITO C S N, GERSTNER W. Nonlinear Hebbian Learning as a Unifying Principle in Receptive Field Formation[C/OL]. [2016-04-22]. http://arxiv.org/pdf/1601.00701v1.pdf. [20] HEGER D, KRISCHER K. Robust Autoassociative Memory with Coupled Networks of Kuramoto-Type Oscillators. Physical Review E, 2016. DOI: 10.1103/PhysRevE.94.022309. [21] GROSSBERG S. Adaptive Pattern Classification and Universal Recoding: II. Feedback, Expectation, Olfaction, Illusions. Biological Cybernetics, 1976, 23(4): 187-202. [22] MARTINETZ T M, BERKOVICH S G, SCHULTEN K J. ′Neural-Gas′ Network for Vector Quantization and Its Application to Time-Series Prediction. IEEE Transactions on Neural Networks, 1993, 4(4): 558-569. [23] KOSKO B. Bi-directional Associative Memory. IEEE Transactions on Systems, Man, and Cybernetics, 1988, 18(1): 49-60. [24] CHEN S, GAO H, YAN W. Improved Exponential Bidirectional Associative Memory. Electronics Letters, 1997, 33(3): 223-224. [25] CHEN S C. Multiple Improved Exponential Bidirectional Associative Memory and Its Decision-Making Performance in Multievidence Reasoning. Chinese Journal of Computers, 2000, 23(11): 1184-1188. [26] ZHANG D Q, CHEN S C. An Extended Multi-valued Exponential Bi-directional Associative Memory Model and Its Application. Journal of Software, 2003, 14(3): 697-702. [27] CHEN L, CHEN S Q, ZHANG D Q. Small World Structure Inspired Many to Many Kernel Associative Memory Models and Their Application. Journal of Software, 2006, 17(1): 223-231. [28] SU M C, LO H H. A Neural Tree with Partial Incremental Learning Capability // Proc of the International Conference on Machine Learning and Cybernetics. Washington, USA: IEEE, 2007: 6-11. [29] 杨海峰,刘 渊,谢振平,等.球向量机的快速在线学习.计算机研究与发展, 2013, 50(9): 1836-1842. (YANG H F, LIU Y, XIE Z P, et al. Efficiently Training Ball Vector Machine in Online Way. Journal of Computer Research and Development, 2013, 50(9): 1836-1842.) [30] 王 涛,刘 渊,谢振平.一种基于飘动性分析的视频烟雾检测新方法.电子与信息学报, 2011, 33(5): 1024-1029. (WANG T, LIU Y, XIE Z P. Flutter Analysis Based Video Smoke Detection. Journal of Electronics & Information Technology, 2011, 33(5): 1024-1029.) [31] VANAMALA S, SREE L P, BHAVANI S D. Rare Association Rule Mining for Data Stream // Proc of the 5th International Conference on Computer and Communications Technologies. Washington, USA: IEEE, 2014. DOI: 10.1109/ICCCT2.2014.7066696. [32] QIAO H, LI Y, TANG T, et al. Introducing Memory and Association Mechanism into a Biologically Inspired Visual Model. IEEE Transactions on Cybernetics, 2014, 44(9): 1485-1496. [33] SUH D I, KIL J P, CHOI Y, et al. An Associative Memory Device Using a Magnetic Tunnel Junction. IEEE Transactions on Magnetics, 2015, 51(11). DOI: 10.1109/TMAG.2015.2444413.
2017, Vol. 30 
