請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/32613完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 劉長遠(Cheng-Yuan Liou) | |
| dc.contributor.author | I-Chun Lin | en |
| dc.contributor.author | 林義淳 | zh_TW |
| dc.date.accessioned | 2021-06-13T04:12:19Z | - |
| dc.date.available | 2006-07-27 | |
| dc.date.copyright | 2006-07-27 | |
| dc.date.issued | 2006 | |
| dc.date.submitted | 2006-07-24 | |
| dc.identifier.citation | [1] R.G. Andrzejak, K. Lehnertz, F. Mormann, C. Rieke, P. David, and C.E. Elger. Indications
of nonlinear deterministic and …nite-dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state. Physical Review E, 64:061907, 2001. [2] A. Babloyantz. Chaotic dynamics in brain activity. dynamics of sensory and cognitive processing by the brain. E. Basar (Eds.), 1998. [3] S. Boccaletti, C. Grebogi, Y.C. Lai, H. Mancini, and D. Mazaet. The control of chaos: Theory and applications. Physics Reports, 329:108–109, 2000. [4] R. Brown, P. Bryant, and H.D.I. Abarbanel. Computing the lyapunov spectrum of a dynamic system from an observed time-series. Physical Review A, 43:2787, 1991. [5] M.C. Casdagli, L.D. Iasemidis, R.S. Savit, R.L. Glimore, S.N. Roper, and J.C. Sackle- lares. Non-linearity in invasive eeg recordings from patients with temporal lobe epilepsy. Eletroencephalogr. Clin. Neurophysiol, 1(102), 1998. [6] A. Das, P. Das, and A. B. Roy. Applicability of lyapunov exponent in eeg data analysis. Complexity International, 9, 2002. [7] A.P. Dempster, N.M Laird, and D.B. Rubin. Maximum likelihood from incomplete data via the em algorithm. J. R. Statist. Soc, 39:1–38, 1977. [8] R.O. Duda and P.E. Hart. Pattern classi…cation and scene analysis. New York: Wiley, 1973. [9] J. P. Eckmann and D. Ruelle. Ergodic theory of chaos and strange attractors. Rev. Mod. Phys., 57:617 ¾ aV656, 1985. [10] J.P. Eckmann, S.O. Kamphorst, D. Ruelle, and S. Ciliberto. Liapunov exponents from time series. Physical Review A, 34:2787, 1986. [11] A. Mees (Ed.). Nonlinear Dynamics and Statistics. Springer, 2001. [12] A.M. FRASER and H.L. Swinney. Independent coordinates for strange attractors from mutual information. Physical Review A, 33:1134, 1986. [13] R. Gencay and W. Dechert. The identi…cation of spurious lyapunov exponents in jacobian algorithms. Studies in nonlinear dynamics and econometrics, 1:145–154, 1996. [14] N. Gershenfeld. Nature of Mathematical Modeling. MIT Press, 1998. [15] N. Gershenfeld, B. Schoner, and E. Metois. Cluster-weighted modeling for time-series analysis. Nature, 397:329–332, 1999. [16] N. Gershenfeld and A.S. Weigend. In: Time series prediction: Forecasting the Future and Understanding the Past ( eds A. S. Weigend and N. A. Gershenfeld). Addison-Wesley, 1993. [17] A. Ghosh and R. Ramaswamy. Cluster-weighted modeling: Estimation of the lyapunov spectrum in driven systems. Physical Review E, 71:016224, 2005. [18] S. Haykin. Neural Networks. Prentice Hall, 1999. [19] R. Hegger, H. Kantz, and T. Schreiber. Practical implementation of nonlinear time series methods: The tisean package. Chaos, 9:413, 1999. [20] M. Jordan and R. Jacobs. Hierarchical mixtures of experts and the em algorithm. Neural Computation, 6:181–214, 1994. [21] M. C. Mackey and L. Glass. Oscillation and chaos in physiological control systems. Science, 197:716–723, 1997. [22] D.F. McCa¤rey, S. Ellner, R. Gallant, and D.W. Nychka. Estimating the lyapunov expo- nent of a chaotic system with nonparametric regression. Journal of the american statistical association, 87:682–695, 1992. [23] D. Nychka, S. Ellner, R. Gallant, and D. MacCa¤rey. Finding chaos in noisy systems. Journal of the Royal statistical society, series B, 54:399–462, 1992. [24] V.I. Oseledec. A multiplicative ergodic theorem. lyapunov characteristic numbers for dy- namical systems. Transactions of the Moscow mathematical society., 19:197–221, 1968. [25] T. Poggio and F. Girosi. Networks for approximation and learning. In Proceedings of the IEEE, volume 78, pages 1481–1497, 1990. [26] M. Rosenstein, J.J. Collins, and C. De Luca. A pratical method for calculating largest lyapunov exponents from small data sets. Physica D, 65:117–134, 1993. [27] T. Sauer, J.A. Yorke, and M. Casdagli. Embedology. Journal of statistical physics, 65:579, 1991. [28] R.H. Shumway and D.S. Sto¤er. Time Series Analysis and ITs Applications. Springer, 2000. [29] R.a.m. van der linden and the sidc team, online catalogue of the sunspot index, 2006. [30] F. Takens. Dynamical systems and turbulence. Lecture notes in mathematics, 898, 1980. [31] J. Theiler. Don’t bleach chaotic data. Chaos, 3:771–782, 1993. [32] M. Thomason. A basic neural network-based trading system: project revisited (part 1 and 2). J. Comput. Intell. Finance, 7(3):36–45, 1999. [33] M. Thomason. A basic neural network-based trading system: project revisited (part 3 and 4). J. Comput. Intell. Finance, 7(4):35–45, 1999. [34] D. Wettschereck and T. Dietterich. Improving the performance of radial basis function net- works by learning center locations. In Advances in Neural Information Processing Systems, volume 4, pages 1133–1140, 1992. [35] A. Wolf, J.B. Swift, H. Swinney, and J. Vastano. Determining lyapunov exponents from a time series. Physica D, 16:285–317, 1985. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/32613 | - |
| dc.description.abstract | 本論文是以叢聚權重模型為基礎, 其模型可以視為一個優良的函數逼近模型. 藉由估計輸入-輸出資料的機率密度來達成. 叢聚權重模型是以期望-最大化 (EM) 演算法來進行訓練. 在本論文中, 最小平方法 (LMS) 被用來更進一步將叢聚權重模型的訓練結果再度訓練, 且可視為一種互補的訓練方法. 因為期望-最大化演算法和最小平方法的目標函數並不相同, 因此兩者的極小值並不會相同. 最小平方法的訓練結果可以用來初始叢聚權重模型的參數, 因此提供了一個可以避免陷入區域極小值的問題. 本論文包含時間序列的預測, 颱風路徑預測以Lyapunov指數的估測實驗. | zh_TW |
| dc.description.abstract | This thesis is based on Cluster-Weighted Modeling (CWM), which can be viewed as a novel uni-versal function approximator based on input-output joint density estimation. CWM is trained by Expectation-Maximization (EM) algorithm. In this thesis Least-Mean-Square (LMS) is ap-
plied to further train the model parameters and it can be viewed as a complementary training method for CWM. Due to different objective functions of EM and LMS, the local minimum should not be the same for the two objective functions. The training result of LMS learning can be used to reinitialize CWM’s model parameters which provides an approach to mitigate local minimum problems. Experiments of time-series prediction, hurricane track prediction and Lyapunov exponents estimation are presented in this thesis. | en |
| dc.description.provenance | Made available in DSpace on 2021-06-13T04:12:19Z (GMT). No. of bitstreams: 1 ntu-95-R93922140-1.pdf: 717286 bytes, checksum: 715934e25daf9cd1d676607bf2af5301 (MD5) Previous issue date: 2006 | en |
| dc.description.tableofcontents | 1 Introduction. . . 8
2 Cluster-Weighted Modeling. . . 10 2.0.1 Architecture. . . 11 2.0.2 Model Estimation . . . 14 3 Least-Mean-Square Training of CWM. . . 22 3.1 Unconstrained Optimization Techniques. . . 22 3.2 Least-Mean-Square Algorithm. . . 24 3.3 Using LMS to Train CWM. . . 24 3.4 Experiments. . . 30 3.4.1 Simulated Data Experiments. . . 30 3.4.2 Real-World Data Experiments. . . 38 3.4.3 Local Minimum. . . 52 4 Using CWM to Estimate Lyapunov Exponents. . . 58 4.1 Lyapunov Exponents. . . 58 4.2 Lyapunov Exponents Estimation. . . 60 4.3 Lyapunov Exponents Estimation of EEG Time-Series Data. . . 62 4.4 Data Collection. . . 62 4.4.1 Results. . . 65 5 Conclusion . . . 66 | |
| dc.language.iso | en | |
| dc.subject | 函數逼近 | zh_TW |
| dc.subject | 叢聚權重模型 | zh_TW |
| dc.subject | 最小平方法 | zh_TW |
| dc.subject | 時間序列 | zh_TW |
| dc.subject | least-mean-square | en |
| dc.subject | function approximation | en |
| dc.subject | time series | en |
| dc.subject | cluster-weighted modeling | en |
| dc.title | 以最小平方法訓練叢聚權重模型 | zh_TW |
| dc.title | Least-Mean-Square Training of Cluster-Weighted Modeling | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 94-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 林智仁(Chih-Jen Lin),程爾觀(Philip E. Cheng),John Aston | |
| dc.subject.keyword | 叢聚權重模型,最小平方法,時間序列,函數逼近, | zh_TW |
| dc.subject.keyword | cluster-weighted modeling,least-mean-square,time series,function approximation, | en |
| dc.relation.page | 69 | |
| dc.rights.note | 有償授權 | |
| dc.date.accepted | 2006-07-26 | |
| dc.contributor.author-college | 電機資訊學院 | zh_TW |
| dc.contributor.author-dept | 資訊工程學研究所 | zh_TW |
| 顯示於系所單位： | 資訊工程學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-95-1.pdf 未授權公開取用 | 700.47 kB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。