A LEARNING ALGORITHM TO TEACH SPATIOTEMPORAL PATTERNS TO RECURRENT NEURAL NETWORKS

被引:37
作者
SATO, M
机构
[1] ATR Auditory and Visual Perception Research Laboratories, Sanpeidani Inuidani, Kyoto, 619-02, Seika-cho, Soraku-gun
关键词
D O I
10.1007/BF00198101
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
A new supervised learning algorithm is proposed. It teaches spatiotemporal patterns to the recurrent neural network with arbitrary feedback connections. In this method the network with fixed connection weights is run for a given period of time under a given external input and initial condition. Then the weights are changed so that the total error from the time dependent teacher signal in this period is maximally decreased. This algorithm is equivalent to the back propagation method for the recurrent network if the discrete time prescription is adopted. However, continuous time formalism seems suited for temporal processing application. © 1990 Springer-Verlag.
引用
收藏
页码:259 / 263
页数:5
相关论文
共 13 条
[1]  
ALMEIDA LB, 1987, 1ST P INT C NEUR NET
[2]   ABSOLUTE STABILITY OF GLOBAL PATTERN-FORMATION AND PARALLEL MEMORY STORAGE BY COMPETITIVE NEURAL NETWORKS [J].
COHEN, MA ;
GROSSBERG, S .
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS, 1983, 13 (05) :815-826
[3]  
Hinton G. E., 1986, PARALLEL DISTRIBUTED, P318
[4]   NEURONS WITH GRADED RESPONSE HAVE COLLECTIVE COMPUTATIONAL PROPERTIES LIKE THOSE OF 2-STATE NEURONS [J].
HOPFIELD, JJ .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA-BIOLOGICAL SCIENCES, 1984, 81 (10) :3088-3092
[5]   NEURAL NETWORKS AND PHYSICAL SYSTEMS WITH EMERGENT COLLECTIVE COMPUTATIONAL ABILITIES [J].
HOPFIELD, JJ .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA-BIOLOGICAL SCIENCES, 1982, 79 (08) :2554-2558
[6]  
IKEDA T, 1988, ITEJ TECH REP, V12, P31
[7]   NETWORK MODEL OF SHAPE-FROM-SHADING - NEURAL FUNCTION ARISES FROM BOTH RECEPTIVE AND PROJECTIVE FIELDS [J].
LEHKY, SR ;
SEJNOWSKI, TJ .
NATURE, 1988, 333 (6172) :452-454
[8]  
Nowlan S. J., 1988, Complex Systems, V2, P305
[9]   Learning State Space Trajectories in Recurrent Neural Networks [J].
Pearlmutter, Barak A. .
NEURAL COMPUTATION, 1989, 1 (02) :263-269
[10]   GENERALIZATION OF BACK-PROPAGATION TO RECURRENT NEURAL NETWORKS [J].
PINEDA, FJ .
PHYSICAL REVIEW LETTERS, 1987, 59 (19) :2229-2232