“你的最高水平就是最近一篇论文”--- 国际学术界流行语

    这句流行语说的是学者应该努力的方向，并不是说这是客观事实。一项研究成果的质量，是由众多因素综合决定的，包括天时地利人和以及作者的努力与能力等等。实际上，那些最具开创性的研究成果，往往是在作者年轻时期做出来的。质量是一个点上的突破，不是面上的扩充。突破一个点，需要的是人类最珍贵的创造力，而通常来讲一个人的创造力在年轻时期是最旺盛的。

    那么，对于一个学者来说，如何在一生的科研之路上努力突破自己、做出更高质量的研究成果呢？我觉得，跨学科研究是突破自己的可行之路。

    跨学科研究不是简单地和其它领域的学者合作，因为现在学科越来越细，两个学科的人说的是不同的语言。突破性的跨学科研究，是一个领域的顶级学者，放下架子，跑到另外一个领域，从博士开始学起，学到另外那个领域的最高点，可以进行创造。也就是说，一个脑子，同时是两个领域的顶级专家，最尖端的思想在一个脑子里碰撞，这样才最有可能做出真正跨学科的突破性成果。

    这也是我自己三十多年来所走的科研之路。下面是我的主要研究成果的简要介绍，从我的座右铭“好好学习，天天向上”开始。

“好好学习，天天向上”

RESEARCH CONTRIBUTIONS: AN INTRODUCTION

For more than twenty years, Li-Xin Wang’s works on fuzzy systems and fuzzy control have been and continue to be very influential. His works are characterized by originality, rigor and easy-to-follow. He initiated four very active research areas in fuzzy systems during the last two decades: adaptive fuzzy control, universal approximation of fuzzy systems, design of fuzzy systems from data, and universal hierarchical fuzzy models, which are briefly summarized as follows.

Although Zadeh proposed fuzzy sets and fuzzy logic as early as 1965, till the early 1990’s fuzzy systems and control were largely rule based expert systems and heuristic control which had limited modelling power, lacking of learning and adaptive capabilities, and unable to use widely available data. The fundamental changes and breakthroughs in this field happened around the early 1990’s, which have led to model based fuzzy systems and control today and Li-Xin Wang is one of the most important researchers who led these fundamental changes and breakthroughs:

·      He is the first researcher proving that fuzzy systems are universal approximators (his two papers for these results – References 4 and 6 in the Record of Accomplishments section below – have been cited 4531 times; Note: All numbers of citations are from Google Scholar as of June 2021), which established the theoretical foundation that fuzzy systems are general nonlinear system models and representations and therefore are widely applicable. This fundamental result has led and created an active research area in fuzzy systems during the last twenty years – fuzzy approximation of general nonlinear systems. 

·      He is one of the first researchers proposing learning or generating fuzzy systems from data (his three papers for these results – References 3, 7 and 11 below – have been cited 4599 times) and this has resulted in and formed the research area – learning and identification of fuzzy systems from data, one of the most important and most active research areas in fuzzy systems during the last two decades. Specifically, the paper Reference 7 below published in the first FUZZ-IEEE conference was one of the three groups showing for the first time how to use the back-propagation algorithm to train a fuzzy system, and the paper Reference 3 below proposed the so-called Wang-Mendel method – the first published method for extracting fuzzy IF-THEN rules from data, which is now the standard against which all other methods are compared.

·      He is the first researcher establishing the basic mathematical framework for adaptive fuzzy systems and control and proving the most fundamental stability and convergence properties. His book published in 1994 on this topic – Reference 1 below – is the most cited book in this area and has been cited 5110 times. His 1993-1996 papers on the subject – References 5, 8 and 10 below – have been cited 2700 times. This book and these papers basically have led and created another most important and active research area in fuzzy systems during the last two decades – adaptive fuzzy systems and control, which has pushed fuzzy control forward from the early stage rule-based expert systems and heuristic control to the more powerful model based fuzzy control for general nonlinear systems which can be analyzed and designed in a mathematically rigorous fashion.

·      He is the first researcher proving that hierarchical fuzzy systems are universal approximators and performing detailed analysis and design of hierarchical fuzzy systems from data (References 9 and 12 below, cited 582 times). Hierarchical fuzzy systems are more powerful modelling structures than standard fuzzy systems, and extending standard fuzzy systems to hierarchical fuzzy systems is similar to the extension of single-hidden-layer neural networks to deep neural networks. With the extraordinary success of deep neural networks, we believe that hierarchical fuzzy systems could also dramatically improve the performance of the fuzzy system approaches to a wide variety of practical problems.

In the educational frontier, his textbook “A Course in Fuzzy Systems and Control” – Reference 2 below – is one of the most widely-used textbooks in the field and has been cited 4938 times.

In recent years, Li-Xin Wang continues his pioneering style of research to explore new frontiers – in finance, economics, and social sciences – for the fuzzy field. Specifically:

·      He creatively used fuzzy systems theory to convert the technical trading heuristics commonly used by stock practitioners into price dynamical equations, studied a number of key problems in financial economics – equilibrium, volatility, predictability and independency – from a new angle, and developed some practically very successful trading strategies based on the fuzzy models (L.X. Wang, “Dynamical models of stock prices based on technical trading rules Part I: The models; Part II: Analysis of the model; Part III: Application to Hong Kong stocks,” IEEE Trans. on Fuzzy Systems 23(4): 787-801 (Part I); 23(4): 1127-1141 (Part II); 23(5): 1680-1697 (Part III), 2015). Specifically, rather than following the traditional approach of concocting stochastic models of market behaviour, his new models are constructive, in the sense that they can be directly incorporated into the design of financial engineering systems, leading to scrutable implementations whose performance can be analyzed in a causal fashion. These results demonstrate to a wider audience the usefulness of fuzzy theory in the study of these very important practical problems.

·      He initiated Fuzzy Opinion Networks – a new mathematical framework for the evolution and propagation of human opinions and their uncertainties across various types of social network structures (L.X. Wang and J.M. Mendel, “Fuzzy opinion networks: A mathematical framework for the evolution of opinions and their uncertainties across social networks,” IEEE Trans. on Fuzzy Systems, Vol. 24, No. 4, pp. 880-905, 2016). He used this new mathematical model to study: (1) the interaction between investor social networks and stock price dynamics (L.X. Wang, “Modeling stock price dynamics with fuzzy opinion networks,” IEEE Trans. on Fuzzy Systems, Vol. 25, No. 2, pp. 277-301, 2017), and (2) the top-down social organizations and the bottom-up election scenarios (L.X. Wang, “Hierarchical fuzzy opinion networks: Top-down for social organizations and bottom-up for election,” IEEE Trans. on Fuzzy Systems, Vol. 28, No. 7, pp. 1265-1275, 2020). This research opened a new door for fuzzy researchers to play an important role in the exciting interdisciplinary fields – social networks and opinion dynamics.

·      He proposed Deep Interpretable Fuzzy Models and fast training algorithms for the new models (L.X. Wang, “Fast training algorithms for deep convolutional fuzzy systems with application to stock index prediction,” IEEE Trans. on Fuzzy Systems, Vol. 28, No. 7, pp. 1301-1314, 2020), which overcome the computational and interpretability problems of the popular black-box and iteratively trained deep neural networks.

Record of Accomplishments

Li-Xin Wang received the Ph.D. degree from the Department of Electrical Engineering, University of Southern California, Los Angeles, CA, USA, in 1992. From 1992 to 1993, he was a Postdoctoral Fellow with the Department of Electrical Engineering and Computer Science, University of California at Berkeley, supported by the research grant of Professor Zadeh. From 1993 to 2007, he was on the faculty of the Department of Electronic and Computer Engineering, The Hong Kong University of Science and Technology (HKUST). In 2007, he resigned from his tenured position at HKUST to become an independent researcher and investor in the stock and real estate markets in China. In Fall 2013, he returned to academia and is now a Professor with University of Chinese Academy of Sciences, Beijing, P.R.China.

His most cited publications are listed next (Note: All numbers of citations are from Google Scholar as of June 2021).

[1] L. X. Wang, Adaptive Fuzzy Systems and Control: Design and Stability Analysis, Prentice-Hall: Englewood Cliffs, NJ, 1994. Referenced 5110 times.

[2] L. X. Wang, A Course in Fuzzy Systems and Control, Prentice-Hall: Englewood Cliffs, NJ, 1997. Referenced 4938 times.

[3] L. X. Wang and J. M. Mendel, “Generating fuzzy rules by learning from examples,” IEEE Trans. on Systems, Man, and Cybern., Vol. 22, No. 6, pp. 1414-1427, 1992. Referenced 3682 times.

[4] L. X. Wang and J. M. Mendel, “Fuzzy basis functions, universal approximation, and orthogonal least squares learning,” IEEE Trans. on Neural Networks, Vol. 3, No. 5, pp. 807-814, 1992. Referenced 3018 times.

[5] L. X. Wang, “Stable adaptive fuzzy control of nonlinear systems,” IEEE Trans. on Fuzzy Systems, Vol. 1, No. 2, pp. 146-155, 1993. Referenced 2017 times.

[6] L. X. Wang, “Fuzzy systems are universal approximators,” Proc. 1992 IEEE International Conf. on Fuzzy Systems, pp. 1163-1170, 1992. Referenced 1513 times.

[7] L. X. Wang and J. M. Mendel, “Back-propagation fuzzy systems as nonlinear dynamic system identifiers,” Proc. 1992 IEEE International Conf. on Fuzzy Systems, pp. 1409-1418, 1992. Referenced 665 times.

[8] L. X. Wang, “Stable adaptive fuzzy controllers, with application to inverted pendulum tracking,” IEEE Trans. on Systems, Man, and Cybern., Vol. 26, No. 5, pp. 677-691, 1996. Referenced 411 times.

[9] L. X. Wang, “Analysis and design of hierarchical fuzzy systems,” IEEE Trans. on Fuzzy Systems, Vol. 7, No. 5, pp. 617-624, 1999. Referenced 334 times.

[10] L. X. Wang, “Design and analysis of fuzzy identifiers of nonlinear dynamic systems,” IEEE Trans. on Automatic Control, Vol. 40, No. 1, pp. 11-23, 1995. Referenced 272 times.

[11] L. X. Wang, “The WM method completed: A flexible fuzzy system approach to data mining,” IEEE Trans. on Fuzzy Systems, Vol. 11, No. 6, pp. 768-782, 2003. (cited times in google scholar = 252)

[12] L. X. Wang, “Universal approximation by hierarchical fuzzy systems,” Fuzzy Sets and Systems, Vol. 93, pp. 223-230, 1998. Referenced 248 times.

RESEARCH INTERESTS

·   Modelling and analysis of asset price dynamics from AI and dynamical system’s perspective.

·   Market microstructure (Limit Order Books) and trading strategies.

·   Deep fuzzy models with big data applications such as on-line advertising and index prediction.

·   Opinion networks and social psychology.

PUBLICATIONS

Recent (2014 — : after I returned to academia in Fall 2013)

Journal Papers:

[1] L. X. Wang, “Fast training algorithms for deep convolutional fuzzy systems with application to stock index prediction,” IEEE Trans. on Fuzzy Systems, Vol. 28, No. 7, pp. 1301-1314, 2020. https://ieeexplore.ieee.org/document/8788632 code：https://ieeexplore.ieee.org/ielx7/91/9130783/8788632/code_training_algorithm.pdf?tp=&arnumber=8788632

[2] L. X. Wang, “Hierarchical fuzzy opinion networks: Top-down for social organizations and bottom-up for election,” IEEE Trans. on Fuzzy Systems, Vol. 28, No. 7, pp. 1265-1275, 2020.  https://ieeexplore.ieee.org/document/8957283

[3] L. X. Wang, “A new look at type-2 fuzzy sets and type-2 fuzzy logic systems,” IEEE Trans. on Fuzzy Systems, Vol. 25, No. 3, pp. 693-706, 2017. (cited times in google scholar = 37)  https://ieeexplore.ieee.org/document/7436787

[4] L. X. Wang, “Modeling stock price dynamics with fuzzy opinion networks,” IEEE Trans. on Fuzzy Systems, Vol. 25, No. 2, pp. 277-301, 2017. (cited times in google scholar = 18)  https://ieeexplore.ieee.org/document/7482744

[5] L. X. Wang and J. M. Mendel, “Fuzzy opinion networks: A mathematical framework for the evolution of opinions and their uncertainties across social networks,” IEEE Trans. on Fuzzy Systems, Vol. 24, No. 4, pp. 880-905, 2016. (cited times in google scholar = 27)  https://ieeexplore.ieee.org/document/7289407

[6] L. X. Wang, “Dynamical models of stock prices based on technical trading rules Part III: Application to Hong Kong stocks,” IEEE Trans. on Fuzzy Systems, Vol. 23, No. 5, pp. 1680-1697, 2015. (cited times in google scholar = 16)  https://ieeexplore.ieee.org/document/6965624

[7] L. X. Wang, “Dynamical models of stock prices based on technical trading rules Part II: Analysis of the models,” IEEE Trans. on Fuzzy Systems, Vol. 23, No. 4, pp. 1127-1141, 2015.  https://ieeexplore.ieee.org/document/6873314

[8] L. X. Wang, “Dynamical models of stock prices based on technical trading rules Part I: The models,” IEEE Trans. on Fuzzy Systems, Vol. 23, No. 4, pp. 787-801, 2015. (cited times in google scholar = 35)  https://ieeexplore.ieee.org/document/6824795

[9] L. X. Wang, “Gaussian-chain filters for heavy-tailed noise with application to detecting big buyers/sellers in stock market,” arXiv:1405.2220, working paper, 2014.  https://arxiv.org/abs/1405.2220

Conference Papers:

[10] L. X. Wang, “Speculative dynamical systems: How technical trading rules determine price dynamics,” Proc. of the 2014 IEEE Symposium Series on Computational Intelligence, Orlando, FL, USA, Dec. 2014.

[11] L. X. Wang, “What happens when trend-followers and contrarians interplay in stock market,” Proc. of the 2014 IEEE Symposium Series on Computational Intelligence, Orlando, FL, USA, Dec. 2014.

[12] L. X. Wang, “How to detect big buyers in Hong Kong stock market and follow them up to make money,” Proc. of the 2014 IEEE Symposium Series on Computational Intelligence, Orlando, FL, USA, Dec. 2014.

[13] L. X. Wang and J. M. Mendel, “Fuzzy networks: What happens when fuzzy people are connected through social networks,” Proc. of the 2014 IEEE Symposium Series on Computational Intelligence, Orlando, FL, USA, Dec. 2014.

Past (1988 — 2006: before I left academia in 2007; listed in descending order according to the number of citations in Google Scholar as of June 2021)

[14] L. X. Wang, Adaptive Fuzzy Systems and Control: Design and Stability Analysis, Prentice-Hall: Englewood Cliffs, NJ, 1994. (cited times in google scholar = 5110)

[15] L. X. Wang, A Course in Fuzzy Systems and Control, Prentice-Hall: Englewood Cliffs, NJ, 1997. (cited times in google scholar = 4938)

[16] L. X. Wang and J. M. Mendel, “Generating fuzzy rules by learning from examples,” IEEE Trans. on Systems, Man, and Cybern., Vol. 22, No. 6, pp. 1414-1427, 1992. (cited times in google scholar = 3682)

[17] L. X. Wang and J. M. Mendel, “Fuzzy basis functions, universal approximation, and orthogonal least squares learning,” IEEE Trans. on Neural Networks, Vol. 3, No. 5, pp. 807-814, 1992. (cited times in google scholar = 3018)

[18] L. X. Wang, “Stable adaptive fuzzy control of nonlinear systems,” IEEE Trans. on Fuzzy Systems, Vol. 1, No. 2, pp.146-155, 1993. (cited times in google scholar = 2017)

[19] L. X. Wang, “Fuzzy systems are universal approximators,” Proc. 1992 IEEE International Conf. on Fuzzy Systems, pp. 1163-1170, 1992. (cited times in google scholar = 1513)

[20] L. X. Wang and J. M. Mendel, “Back-propagation fuzzy systems as nonlinear dynamic system identifiers,” Proc. 1992 IEEE International Conf. on Fuzzy Systems, pp. 1409-1418, 1992. (cited times in google scholar = 665)

[21] L. X. Wang, “Stable adaptive fuzzy controllers, with application to inverted pendulum tracking,” IEEE Trans. on Systems, Man, and Cybern., Vol. 26, No. 5, pp. 677-691, 1996. (cited times in google scholar = 411)

[22] L. X. Wang, “Analysis and design of hierarchical fuzzy systems,” IEEE Trans. on Fuzzy Systems, Vol. 7, No. 5, pp. 617-624, 1999. (cited times in google scholar = 334)

[23] L. X. Wang, “Design and analysis of fuzzy identifiers of nonlinear dynamic systems,” IEEE Trans. on Automatic Control, Vol. 40, No. 1, pp. 11-23, 1995. (cited times in google scholar = 272)

[24] L. X. Wang, “The WM method completed: A flexible fuzzy system approach to data mining,” IEEE Trans. on Fuzzy Systems, Vol. 11, No. 6, pp. 768-782, 2003. (cited times in google scholar = 252)

[25] L. X. Wang, “Universal approximation by hierarchical fuzzy systems,” Fuzzy Sets and Systems, Vol. 93, pp. 223-230, 1998. (cited times in google scholar = 248)

[26] L. X. Wang and J. M. Mendel, “Fuzzy adaptive filters, with application to nonlinear channel equalization,” IEEE Trans. on Fuzzy Systems, Vol.1, No.3, pp. 161-170, 1993. (cited times in google scholar = 195)

[27] L. X. Wang, “A supervisory controller for fuzzy control systems that guarantees stability,” IEEE Trans. on Automatic Control, Vol. 39, No. 9, pp. 1845-1848, 1994. (cited times in google scholar = 148)

[28] L. X. Wang, “Stable and optimal fuzzy control of linear systems,” IEEE Trans. on Fuzzy Systems, Vol. 6, No. 1, pp. 137-143, 1998. (cited times in google scholar = 117)

[29] L. X. Wang and J. M. Mendel, “Adaptive minimum prediction-error deconvolution and source wavelet estimation using Hopfield neural networks,” Geophysics, Vol. 57, No. 5, pp. 670-679, 1992. https://library.seg.org/doi/10.1190/1.1443281 (cited times in google scholar = 97)

[30] L. Zhou, C. Chang and L. X. Wang, “Adaptive fuzzy control for nonlinear building-magnetorheological damper system.” J. Struct. Eng. 129(7): 905–913, 2003. (cited times in google scholar = 93)

[31] L. X. Wang, “Training of fuzzy logic systems using nearest neighborhood clustering,” Proc. of 1993 IEEE International Conf. on Fuzzy Systems, pp. 13-17, San Francisco, March 1993. (cited times in google scholar = 90)

[32] C. Wei and L. X. Wang, “A note on universal approximation by hierarchical fuzzy systems,” Information Sciences, Vol. 123, pp. 241-248, 2000. (cited times in google scholar = 89)

[33] L. X. Wang and F. Wan, “Structured neural networks for constrained model predictive control,” Automatica, Vol. 37, pp. 1235-1243, 2001. (cited times in google scholar = 76)

[34] L. X. Wang and J. M. Mendel, “Structured trainable networks for matrix algebra,” Proc. of 1990 International Joint Conference on Neural Networks, pp.  125-132, 1990. (cited times in google scholar = 73)

[35] F. L. Lewis, K. Liu, R. Selmic and L. X. Wang, “Adaptive fuzzy logic compensation of actuator deadzones,” Journal of Robotic Systems, Vol. 14, No. 6, pp. 501-511, 1997. (cited times in google scholar = 69)

[36] L. X. Wang, “Modeling and control of hierarchical systems with fuzzy systems,” Automatica, Vol. 33, No. 6, pp. 1041-1053, 1997. (cited times in google scholar = 59)

[37] L. X. Wang and C. Wei, “Approximation accuracy of some neuro-fuzzy approaches,” IEEE Trans. on Fuzzy Systems, Vol. 8, No. 4, pp. 470-478, 2000. (cited times in google scholar = 58)

[38] L. X. Wang and J. M. Mendel, “Three-dimensional structured networks for matrix equation solving,” IEEE Trans. on Computers, Vol. 40, No. 12, pp. 1337-1346, 1991. (cited times in google scholar = 54)

[39] F. L. Lewis, W. K. Tim, L. X. Wang and Z. X. Li, “Deadzone compensation in motion control systems using adaptive fuzzy logic control,” IEEE Trans. on Control Systems Technology, Vol. 7, No. 6, pp. 731-742, 1999. (cited times in google scholar = 52)

[40] F. Wan, H. Shang, L. X.Wang and Y. X. Sun, “How to determine the minimum number of fuzzy rules to achieve given accuracy: A computational geometric approach to SISO case,” Fuzzy Sets and Systems, Vol. 150, pp. 199-209, 2005. (cited times in google scholar = 50)

[41] L. X. Wang and J. M. Mendel, “Parallel structured networks for solving a wide variety of matrix algebra problems,” Journal of Parallel and Distributed Computing, Vol. 14, pp. 236-247, 1992. (cited times in google scholar = 49)

[42] K. T. Woo, L. X. Wang, F. L. Lewis and Z. X. Li, “A fuzzy system compensator for backlash,” Proc. of the 1998 IEEE International Conference on Robotics and Automation, Vol. 1, pp. 181-186, Leuven, may 1998.  (cited times in google scholar = 38)

[43] L. X. Wang, “Fuzzy systems as nonlinear dynamic system identifiers I: Design,” Proc. of the 31^st IEEE Conf. on Decision and Control, Vol.1, pp. 897-902, Tucsun, Dec. 1994. (cited times in google scholar = 34)

[44] L. X. Wang, “Combining mathematical model and heuristics into controllers: an adaptive fuzzy control approach,” Fuzzy Sets and Systems, Vol. 89, pp. 151-156, 1997. (cited times in google scholar = 27)

[45] L. X. Wang, “Automatic design of fuzzy controllers,” Automatica, Vol. 35, pp. 1471-1475, 1999. (cited times in google scholar = 24)

[46] L. X. Wang and J. M. Mendel, “Cumulant-based parameter estimation using structured networks,” IEEE Trans. on Neural Networks, Vol. 2, No. 1, pp. 73-83, 1991. (cited times in google scholar = 23)

[47] L. X. Wang and J. M. Mendel, “An RLS fuzzy adaptive filter, with application to nonlinear channel equalization,” Proc. of 1993 IEEE Conf. on Fuzzy Systems, pp. 895-900, San Francisco, March 1993. (cited times in google scholar = 23)

[48] L. X. Wang, “Solving fuzzy relational equations through network training,” Proc. of 1993 IEEE International Conf. on Fuzzy Systems, pp. 956-960, San Francisco, March 1993. (cited times in google scholar = 19)

[49] L. X. Wang, “Fuzzy systems: Challenges and opportunities,” ACTA Automatica Sinica (in Chinese), Vol. 27, No. 4, pp. 585-590, 2001. (cited times in google scholar = 19)

[50] K. P. Cheung and L. X. Wang, “Comparison of fuzzy and PI controllers for a benchmark drum-boiler model,” Proc. of 1998 IEEE Conf. on Control Applications, Trieste, Italy, Sept. 1998. (cited times in google scholar = 16)

[51] L. X. Wang and J. M. Mendel, “A fuzzy approach to hand-written rotation-invariant character recognition,” Proc. ICASSP-92, 1992. (cited times in google scholar = 15)

[52] L. X. Wang and J. M. Mendel, “Matrix computations and equation solving using structured networks and training,” Proc. 1990 IEEE Conf. on Decision and Control, pp. 1747-1750, 1990. (cited times in google scholar = 15)

[53] L. X. Wang, “A mathematical formulation of hierarchical systems using fuzzy logic systems,” Proc. IEEE World Congress on Computational Intelligence, Florida, June 1994.

[54] L. X. Wang, “A neural detector for seismic reflectivity sequences,” IEEE Trans. on Neural Networks, Vol. 3, No. 2, pp. 338-340, 1992.

[55] F. Wan, H. Shang and L. X. Wang, “Fuzzy model based adaptive predictive control of nonlinear  systems: One-step-ahead and multi-step-ahead scheme,” Proc. Of the IEEE 2004 Conf. On Decision and Control, Atlantis, Bahamas, Dec. 2004.

[56] F. Wan, H. Shang, L. X. Wang and Y. X. Sun, “Neutralization process control using an adaptive fuzzy controller,” Proc. Of the 2004 American Control Conf., Boston, June 2004.

[57] L. X. Wang and F. Wan, “Adaptive Minimum Prediction-Error Fuzzy Control of General Nonlinear Systems”, The IEEE International Conference on Fuzzy Systems, St. Louis, MO, USA, May 25-28, 2003

[58] L. X. Wang and F. Wan, “Identification of higher-order nonlinear dynamic systems with hierarchical fuzzy models,” The 2003 Chinese Control Conference, Yichang, China, July 2003.

[59] F. Wan and L. X. Wang, “One-step-ahead adaptive fuzzy controller for a general class of nonlinear systems,” American Control Conference, Arlington, Virginia, June 2001.

[60] F. Wan and L. X. Wang, “Generating persistently exciting inputs for nonlinear dynamic system identification using fuzzy models,” IEEE International Conf. on Fuzzy Systems, Melbourne, Dec. 2001.

[61] F. Wan and L. X. Wang, “How to determine the minimum number of rules to achieve given accuracy,” IEEE International Conf. on Fuzzy Systems, Melbourne, Dec. 2001.

[62] L. X. Wang, “A fuzzy projection pursuit model for prediction in high-dimensional space with fuzzily dependent inputs,” IEEE International Conf. on Fuzzy Systems, Melbourne, Dec. 2001.

[63] F. Wan and L. X. Wang, “On the persistent excitation conditions of adaptive fuzzy systems,” IEEE Conf. on Decision and Control, Sydney, Dec. 2000.

[64] F. Wan and L. X. Wang, “Design of economical fuzzy systems,” IEEE Conf. on Fuzzy Systems, Texas, May 2000.

[65] L. X. Wang and C. Wei, “Discrete-time adaptive fuzzy control of first-order continuous-time nonlinear systems,” IEEE Conf. on Decision and Control, Phoenix, Arizona, Dec. 1999.

[66] C. Wei and L. X. Wang, “Analysis of the table look-up scheme and clustering method for designing fuzzy systems from data,” 14^th IFAC World Congress, Beijing, July 1999.

[67] K.P. Cheung and L. X. Wang, “Fuzzy system tuned PI controller for a benchmark drum-boiler model,” Proc. of 1998 IEEE Conf. on Decision and Control, Tampa, USA, Dec. 1998.

[68] L. X. Wang, “Designing fuzzy models for nonlinear discrete-time systems with guaranteed performance,” Proc. of 1998 American Control Conference, Philadelphia, June 1998.

[69] L. X. Wang, “Automatic design of fuzzy controllers,” Proc. of 1998 American Control Conference, Philadelphia, June 1998.

[70] L. X. Wang, “A fuzzy system compensator for backlash,” Proc. of the Third Joint Conf. on Information Sciences, Duke University, March 1997.

[71] L. X. Wang, “Universial approximation by hierarchical fuzzy systems,” Proc. of 35th IEEE Conf. on Decision and Control, Japan, Dec. 1996.

[72] L. X. Wang, “Stable and optimal fuzzy control of linear systems,” Proc. of 5th IEEE International Conf. on Fuzzy Systems, Vol. 2, pp. 1453-1458, New Orleans, 1996.

[73] L. X. Wang, “Modeling and control of hierarchical systems with fuzzy systems,” Proc. of 13^th IFAC World Congress, Vol. F, pp. 109-114, San Francisco, 1996.

[74] L. X. Wang, “Modeling and control of hierarchical systems II: controller design,” Proc. of 1995 American Control Conference, Seattle, June 1995.

[75] L. X. Wang, “Design of adaptive fuzzy controllers for nonlinear systems by input-output linearization,” Proc. of the NAFIPS/IFIS/NASA'94\fR, Texas, Dec. 1994.

[76] L. X. Wang, “Stable adaptive fuzzy control of nonlinear systems,” Proc. 31st IEEE Conf. on Decision and Control, pp. 2511-2516, 1992.

[77] L. X. Wang, “Fuzzy systems as nonlinear dynamic system identifiers part II: stability analysis and simulations,” Proc. 31st IEEE Conf. on Decision and Control, pp. 3418-3422, 1992.

[78] L. X. Wang and J. M. Mendel, “Adaptive prediction-error deconvolution and wavelet estimation using Hopfield neural networks,” Proc. ICASSP-91, 1991.

[79] L. X. Wang and J. M. Mendel, “Generating fuzzy rules by learning from examples,” Proc. 6^th IEEE International Symp. on Intelligent Control, pp. 263-268, 1991.

[80] L. X. Wang and J. M. Mendel, “Three-dimensional structured networks for matrix algebra,” Proc. 1991 IEEE Workshop on Neural Networks for Signal Processing, 1991.

[81] L. X. Wang and J. M. Mendel, “Cumulant-based parameter estimation using neural networks,” Proc. ICASSP-90, 1990.

[82] L. X. Wang, G. Z. Dai and J. M. Mendel, “One-pass minimum variance deconvolution algorithms,” IEEE Trans. on Automatic Control, Vol. 35, No. 3, pp. 326-329, 1990.

[83] L. X. Wang and G. Z. Dai, “Distributed estimation for team decision systems,” Decision and Control (in Chinese), Vol. 3, No. 3, pp. 1-7, 1988.

[84] L. X. Wang and G. Z. Dai, “A distributed target tracking algorithm for multiple stations,” Acta Aeronautica et Astronautica Sinica (in Chinese), Vol. 9, No. 3, pp. 165-170, 1988.

[85] L. X. Wang and G. Z. Dai, “A review of distributed estimation theory,” Control Theory and Applications (in Chinese), Vol. 5, No. 1, pp. 1-11, 1988.

[86] L. X. Wang and G. Z. Dai, “New recursive smoothing algorithms for Bernoulli-Gaussian input sequences,” Proc. IFAC 1988 Symp. on Identification and System Parameter Estimation, 1988.

[87] L. X. Wang and G. Z. Dai, “A systematic approach to human properties in complex systems,” Proc. 1988 IEEE International Conf. on  Systems, Man, and Cybern., pp. 1253-1256, 1988.