吕宝粮


教授、博导



所属系别: 计算机系

研究方向: 仿脑计算机理论与模型、神经网络理论与应用、机器学习、人脸检测与识别、自然语言处理、脑与计算机接口、计算系统生物学。

联系电话: 02162932287

E-Mail: blu@cs.sjtu.edu.cn

实验室主页: http://bcmi.sjtu.edu.cn



个人简介
教育及工作经历: 工学博士、教授、博士生导师、IEEE高级会员。1960年11月生于青岛。1982年1月毕业于青岛科技大学自动化系，获工学学士。同年留校任教。1989年4月毕业于西北工业大学计算机科学与工程系，获工学硕士学位。1991年4月至1994年3月在日本京都大学电气工程系攻读博士学位。主要从事模块化神经网络结构与学习算法和多层神经网络逆映像的计算方法及其应用的研究，提出了多级筛选神经网络模型和基于线性与非线性规划方法的多层神经网络逆映像计算方法。1994年3月获京都大学工学博士学位。1994年4月至1999年3月在日本理化学研究所仿生物控制研究中心任研究员，主要参与日本国家重点研究课题“仿生物控制与自律分散系统”的研究，负责“大规模、复杂模式识别问题的分解与学习”子课题，提出了基于类关系的通用问题分解方法和并列模块化神经网络模型（Min-Max Modular Neural Network）。该模型解决了传统多层前馈网络和反向传播学习算法在解决大规模实际问题时所存在的陷于局部极小值、长时间学习和网络结构设计等问题。该模型已成功地应用于脑电波信号分类、自然语言处理中的自动词性标注、和大规模词汇库的自动纠错等问题。1999年4月至2002年8月在日本理化学研究所脑科学综合研究中心任研究员，主要参与日本国家重点研究课题“创造脑”的研究，负责“仿脑计算机的结构与超并列学习模型”子课题。提出了涌现学习方法、具有局部响应的高斯零交叉判别函数和基于涌现学习方法的仿脑计算机模型。2002年8月起任上海交通大学计算机科学与工程系教授，同年12月被评为博士生导师。

社会及学术兼职:

获奖情况:


教学与科研
主讲课程: 1. 算法设计与数据结构
2. 神经网络理论与应用

主要研究领域: 1. 仿脑计算机理论与模型
2. 神经网络理论与应用
3. 机器学习
4. 人脸检测与识别
5. 计算系统生物学
6. 脑-计算机接口
7. 自然语言处理

科研项目: 1. 国家自然科学基金项目：增量学习模型研究；2004.1-2006.12 (项目编号：60375022）
2. 国家自然科学基金项目：超并列模式分类器的问题分解与模块集成研究；2005.1-2007.12 (项目编号:60473040)
3. 中日合作研究项目：基于增量学习模型的图像错误诊断系统；2004.6-2005.3
4. 中日合作研究项目：基于超并列机器学习方法的大规模文本分类研究；2004.4-2005.3

发表论文: Book Chapters:

1) B. L. Lu and K. Ito, ``Transformation of nonlinear programming problems into separable ones using multi-layer neural networks", Mathematics of Neural Networks: Models, Algorithms and Applications, S. W. Ellacott, J. C. Mason, and I. J. Anderson Eds., Kluwer Academic Publishers, pp. 235-239, 1997

Journal Papers:

1) B. L. Lu, J. Shin, and M. Ichikawa, “Massively parallel classification of single-trial EEG signals using a min-max modular neural network”, IEEE Trans. Biomedical Engineering, vol. 51, no. 3, pp. 551-558, 2004
2) B. L. Lu and K. Ito, “Converting general nonlinear programming problems into separable programming problems with feedforward neural networks”, Neural Networks, vol. 16, pp. 1059-1074, 2003
3) B. L. Lu, Q. Ma, M. Ichikawa, and H. Isahara, “Efficient part-of-speech tagging with a min-max modular neural network”, Applied Intelligence, vol 19, pp. 65-81, 2003
4) Q. Ma, B. L. Lu, H. Isahara, and M. Ichikawa, ``Part of speech tagging with min-max modular neural networks”, Systems and Computers in Japan, vol. 33, no. 7, 2002
5) B. L. Lu, J. Shin, and M. Ichikawa, “Effects of features on generalization accuracy of min-max modular neural networks in the classification of single-trial EEG signals”, RIKEN Review, vol. 48, 2002
6) J. Shin, B. L. Lu, A. Talnov, G. Matsumoto, and J. Brankack, “Reading auditory discrimination behaviour of freely moving rats from hippocampal EEG”, Neurocomputing, vol 38-40, pp. 1557-1566, 2001
7) B. L. Lu, J. Shin, and M. Ichikawa, “Fast classification of high-dimensional EEG signals using min-max modular neural networks”, RIKEN Review, vol. 40, pp. 58-62, 2001
8) Q. Ma, B. L. Lu, H. Isahara, and M. Ichikawa, “part of speech tagging with min-max modular neural networks” (in Japanese), IEICE Transactions on Information and Systems, D-II, vol. J84-D-II, no. 4, pp. 708-717, 2001
9) B. L. Lu, Q. Ma, M. Ichikawa, and H. Isahara, “Massively parallel learning of part-of-speech disambiguation”, RIKEN Review, no. 30, pp. 40-49, 2000
10) B. L. Lu, H. Kita, and Y. Nishikawa, “Inverting feedforward neural networks using linear and nonlinear programming”, IEEE Trans. Neural Networks, vol. 10, no. 6, pp. 1271-1290, 1999
11) B. L. Lu and M. Ito, “Task decomposition and module combination based on class relations: a modular neural network for pattern classification”, IEEE Trans. Neural Networks, vol. 10, no. 5, pp. 1244-1256, 1999
12) B. L. Lu and K. Ito, ``Computation of multiple inverse kinematic solutions for redundant manipulators by inverting modular neural networks", Transactions of Institute of Electrical Engineers of Japan, vol. 116-C, No. 1, pp. 49-56, 1996

International Conference Papers:

1. B. L. Lu, K. A. Wang, and Y. M. Wen, “Comparison of parallel and cascade methods for training support vector machines on large-scale problems” (invited paper), Proc. of International Conference on Machine Learning and Cybernetics ( ICMLC04), pp. 3056-3061， Shanghai, China, Aug. 26-29, 2004
2. Y. M. Wen and B. L. Lu, “A cascade method for reducing training time and the number of support vectors”, Advances in Neural Networks-ISNN2004, Lecture Notes in Computer Science, vol. 3173, part I, pp. 480-485, 2004
3. Z. G. Fan and B. L. Lu, “An adjusted gaussian skin-color model based on principle component analysis”, Advances in Neural Networks-ISNN2004, Lecture Notes in Computer Science, vol. 3173, part I, pp. 804-809, 2004
4. H. Zhao and B. L. Lu, “Analysis of fault tolerance of a combining classifier”, Advances in Neural Networks-ISNN2004, Lecture Notes in Computer Science, vol. 3173, part I, pp. 888-893, 2004
5. B. L. Lu, K. A. Wang, M. Utiyama, and H. Isahara, “A part-versus-part method for massively parallel training of support vector machines”, Proc. of IEEE/INNS Int. Joint Conf. on Neural Networks ( IJCNN2004), pp.735-740, Budabest, Hungary, July 25-29, 2004
6. B. L. Lu, “A massively parallel machine learning approach to text categorization”, The 3rd Japan-China Natural Language Processing Joint Research Promotion Conference (invited paper), Shiga, Japan, Nov. 11, 2003
7. B. L. Lu and M. Ichikawa, “Emergent on-line learning with a Gaussian zero-crossing discriminant function”, Proc. of IEEE/INNS Int. Joint Conf. on Neural Networks, Honolulu, USA, pp. 1263-1268, 2002
8. B. L. Lu and M. Ichikawa, “Emergent on-line learning in min-max modular neural networks”, Proc. of IEEE/INNS Int. Conf. on Neural Networks, Washington DC, USA, pp. 2650-2655, 2001
9. B. L. Lu, J. Shin, and M. Ichikawa, “Massively parallel classification of EEG signals using min-max modular neural networks”, Lecture Notes in Computer Science, vol. 2130, pp. 601-608, 2001
10. Q. Ma, B. L. Lu, M. Murata, H. Isahara, and M. Ichikawa, “On-line error detection of annotated corpus using modular neural networks”, Lecture Notes in Computer Science, vol. 2130, pp. 1185-1192, 2001
11. B. L. Lu and M. Ichikawa, “Emergent on-line learning: towards brain-style computer”, Proc. of Int. Conf. on Neural Information Processing , Shanghai, China, vol. 2, pp. 655-658, 2001
12. Q. Ma and B. L. Lu, “Emergent learning and natural language processing”, Proc. of Int. Conf. on Neural Information Processing, Shanghai, China, vol. 2, pp. 659-664, 2001
13. B. L. Lu and M. Ichikawa, “A Gaussian zero-crossing discriminat function for min-max modular neural networks”, in Proc. of 5th International Conference on Knowledge-Based Intelligent Information Engineering Systems & Allied Technologies (KES’01), pp. 298-302, Osaka, Japan, 6-8 September, 2000
14. B. L. Lu and M. Ichikawa, “Emergence of learning: an approach to coping with NP-complete problems in learning”, Proc. of IEEE-INNS-ENNS Int. Joint Conf. On Neural Networks, Como, Italy, vol. 4, pp. 159-164, 2000
15. B. L. Lu, M. Ichikawa and S. Hosoe, “A modular massively parallel learning framework for brain-like computers”, Proc. 1999 IEEE Int. Conf. On Systems, Man and Cybernetics, pp. 332-337, 1999
16. Q. Ma, B. L. Lu, H. Isahara, “Part of speech tagging with min-max modular neural networks”, Proc. 1999 IEEE Int. Conf. On Systems, Man and Cybernetics, pp. 356-360, 1999
17. G. J. Ji, B. L. Lu, X. Chen, and J. Wang, “Object searching in scale-space”, Proc. 1999 IEEE Int. Conf. On Systems, Man and Cybernetics, pp. 356-360, 1999
18. B. L. Lu, Q. Ma, M. Ichikawa, and H. Isahara, “Massively parallel part of speech tagging using min-max modular neural networks”, Proc. of 1st Workshop on Natural Language Processing and Neural Networks, pp. 58-63, 1999
19. B. L. Lu and M. Ito, ``Decomposing and parallel learning of large-scale pattern recognition problems using min-max modular neural network", Proc. of International ICSC/IFAC Symposium on Neural Computation, Vienna, Austria, pp. 703-709, 1998
20. B. L. Lu and M. Ito, ``Decomposing and parallel learning of Imbalanced Classification problems by min-max modular neural network", Proc. of International Conference on Neural Information Processing, Kitakyushu, Japan, pp. 199-202, 1998
21. B. L. Lu and M. Ito, ``Task decomposition based on class relations: a modular neural network architecture for pattern classification", Biological and Artificial Computation: From Neuroscience to Technology, Lecture Notes in Computer Science, J. Mira, R. Moreno-Diaz and J. Cabestany, Eds., vol. 1240, pp. 330-339, Springer, 1997
22. B. L. Lu and K. Ito, ``Solving inverse kinematics problem of redundant manipulators in an environment with obstacles using separable nonlinear programming", Proc. of 1996 Japan-USA Symposium on Flexible Automation, Boston, pp. 79-82, 1996
23. B. L. Lu and K. Ito, ``A parallel and modular multi-sieving neural network architecture with multiple control networks", Proc. of IEEE International Conference on Systems, Man and Cybernetics, Beijing, pp. 1303-1308, 1996
24. B. L. Lu, K. Ito, H. Kita, and Y. Nishikawa, ``A parallel and modular multi-sieving neural network architecture for constructive Learning", Proc. of IEE International Conference on Artificial Neural Networks, Cambridge, UK, pp.92-97, 1995
25. B. L. Lu and K. Ito, ``Regularization of inverse kinematics for redundant manipulators using neural network inversions", Proc. of IEEE International Conference on Neural Networks, Perth, pp. 2726-2731, 1995
26. B. L. Lu, H. Kita, and Y. Nishikawa, ``A multi-sieving neural network architecture that decomposes learning tasks automatically", Proc. of IEEE International Conference on Neural Networks, Orlando, FL, pp.~1319-1324, 1994
27. B. L. Lu, Y. Bai, H. Kita, and Y. Nishikawa, ``An efficient multiplayer quadratic perceptron for pattern classification and function approximation"", Proc. of International Joint Conference on Neural Networks, Nagoya, vol. 2, pp. 1385-1388, 1993
28. B. L. Lu, H. Kita, and Y. Nishikawa, “A new method for inverting nonlinear multiplayer feedforward networks”, Proc. of IEEE International Conference on Industrial Electronics, Control and Instrumentation, Kobe, pp. 1349-1354, 1991

Technical Reports and Oral Presentation:

1) 吕宝粮 王逸飞. 一种基于涌现理论的增量学习模型. 第十三届全国神经网络学术大会论文集. pp. 213-218, 人民邮电出版社, 2003
2) B. L. Lu and M. Ichikawa, “An Emergent learning method capable of training a class of pattern classifiers in polynomial time and space”, Technical Report of IEICE, July 26, 2002
3) B. L. Lu and M. Ichikawa, “Emergent learning; from linear threshold gates to universal classifiers”, Int. Workshop on Information Processing on the Brain 2000, Hangzhu, China, 2000
4) B. L. Lu and M. Ichikawa, “Coping with NP-complete problems in learning of modular neural networks”, Technical Report of IEICE, NC99-150-181, pp. 173-179, 2000
5) B. L. Lu and M. Ichikawa, “An upper bound on the size of min-max modular neural networks trainable in polynomial time”, Technical Report of IEICE, NLP2000-15-29, pp. 55-62, 2000
6) Q. Ma, B. L. Lu, H. Isahara, and M. Ichikawa, ``Disambiguation of part-of-speech with modular neural networks" (in Japanese), Technical Report of IEICE, NC-2000-17, pp. 63-70, 2000
7) B. L. Lu and M. Ito, ``A modular neural network architecture for pattern classification II: recursive decomposing and parallel learning of large-scale problems", Technical Report of IEICE, NC-97-149, pp. 79-86, 1998
8) B. L. Lu and M. Ito, ``A modular neural network architecture for pattern classification III: massively parallel learning", Technical Report of IEICE, NC-98-81, pp. 25-32, 1998
9) B. L. Lu, ``A massively parallel and modular framework for brain-like computers", Technical Report of IEICE, NC-98-140, pp. 327-334, 1998.
10) B. L. Lu and M. Ito, ``A modular neural network architecture for pattern classification I: task decomposition and module combination based on class relations", Technical Report of IEICE, NC-96-158, pp. 23-30, 1997
11) B. L. Lu and M. Ito, “Parallel learning for pattern classification problems using modular neural networks: the case of DNA sequences", Proc. of the 41th Annual Conference of the Institute of Systems, Control and Information Engineers, pp. 69-70, 1997
12) B. L. Lu and M. Ito, ``Space grid partition learning: an alternative modular neural network model for pattern classification", Proc. of the 9th SICE Symposium on Autonomous Decentralized Systems, pp. 199-204, 1997
13) B. L. Lu and K. Ito, ``A comparative study of output representation schemes for multi-layer neural networks", Proc. of 40th Annual Conference of the Institute of Systems, Control and Information Engineers, pp. 1535-1538, 1996
14) B. L. Lu and K. Ito K, ``Transformation of nonlinear programming problems into separable ones using multi-layer neural networks", Proc. 1996 Annual Conference of Japanese Neural Network Society, pp. 336-337, 1996
15) B. L. Lu, ``Optimal inverse kinematics solutions for partial fault redundant manipulators using nonlinear programming", Proc. of 14th Annual Conference of Robotics Society of Japan, pp. 304-305, 1996
16) B. L. Lu and K. Ito, ``Hierarchical computation of optimal Inverse kinematic solution using neural network inversions", Proc. of 6th SICE Symposium on Autonomous Decentralized Systems, pp. 23-28, 1995
17) B. L. Lu and K. Ito, `` Scaling reinforcement learning to autonomous robot action control by decomposing tasks", Proc. of 1995 Annual Joint Conference on Electrical Engineering, pp. 29-30, 1995
18) B. L. Lu, H. Kita, and Y. Nishikawa, ``Learning by multi-sieving neural networks", Proc. of 5th SICE Symposium on Autonomous Decentralized Systems, pp.71-76, 1994
19) B. L. Lu, K. Ito, H. Kita, and Y. Nishikawa, ``A parallel and modular multi-sieving neural network architecture for pattern classification", Proc. of 4th Symposium on Intelligent Systems, pp. 401-406, 1994
20) B. L. Lu, K. Ito, H. Kita, and Y. Nishikawa, ``A parallel multi-sieving neural network architecture", Proc. of 1994 Annual Conference of Japanese Neural Network Society, pp. 237-238, 1994
21) B. L. Lu and Y. Nishikawa, ``A feedforward neural network with quadratic units"", Proc. of the 4th SICE Symposium on Autonomous Decentralized Systems, pp. 33-38, 1993
22) B. L. Lu, Y. Bai, H. Kita, and Y. Nishikawa, ``On learning speed of linear and quadratic perceptron"", Proc. of 1993 Annual Conference of Japanese Neural Network Society, pp. 118-119, 1993
23) B. L. Lu, H. Kita, and Y. Nishikawa, ``Comparison of performances of the first-order and the second-order multi-layer perceptrons", Proc. of the 37th Annual Conference of the Institute of Systems, Control and Information Engineers, pp. 19-20, 1993
24) B. L. Lu, H. Kita, and Y. Nishikawa, ``Examining generalization by network inversions"", Proc. of 1991 Annual Conference of Japanese Neural Network, pp. 27-28, 1991

专利:

其他: