Learning Algorithms and Fixed Dynamics

  title={Learning Algorithms and Fixed Dynamics},
  author={Peter R. Conwell}
A hstracr-This paper discusses the equivalence of learning algorithms and nonlinear dynamic systems whose differential equations have fixed coefficients. We show how back propagation transforms into a fixed-weight recursive neural network suitable for VLSI or optical implementations. The transformation is quite general a n d implies that understanding physiological networks may require one to determine the values of fixed parameters distributed throughout a network. Equivalently, a particular… CONTINUE READING