| 摘要 |
A computing element for use in an array in a neural network. Each computing element has K (K>1) input signal terminals, K input backpropagated signal terminals, K output backpropagated signal terminals and at least one output terminal. The input terminals of the computing element located in row i, column j of the array of computing elements receive a sequence of concurrent input signals on K parallel input lines representing a parallel input signal Sij having vector elements (sij1, sij2, sij3, . . . , sijk)T. The K input backpropagated signal terminals are coupled to receive an m-dimensional (m<K) backpropagated signal vector characterized to provide a measure of the performance error of the computing element. The computing element comprises a weighting function means responsive to the concurrent input signal Sij for computing a K-dimensional weighting coefficient and a scalar activation signal uij by computing a K-dimensional weighting-coefficient vector, Wij=(wij1, wij2, . . . , wijk)T, and by forming the vector inner product of the input signal Sij vector elements and the weighting-coefficient vector, Wij. Feedback signals xijk=wijk*pij are provided from the output backpropagated signal terminals where pij provides a performance error of all columns of computing elements subsequent to this computing element weighted by the gain of the computing element. A nonlinear processor maps successive values of uij through a nonlinear mapping function, Mij to provide a single-valued output signal yij. The nonlinear processor responds to an m-dimensional backpropagated signal vector, Xij+1=(x1,j+1,p; x2,j+1,p; x3,j+1,p; . . . xm,j+1,p)T characterized to provide a measure of the performance error of all computing elements subsequent to the computing element. Vector Xi,j+1 comprises the pth member of all backpropagated error vectors from the computing elements of column j+1.
|
