next up previous contents index
Next: #pollack90j###: Up: Papers Previous: #jordan86t###:   Contents   Index

Elman (1990):

Elman's paper , ``Finding structure in time'' (http://www.dlsi.ua.es/~mlf/nnafmc/papers/elman90finding.pdf), introduces another widely-used recurrent architecture, the simple recurrent net, which everyone calls now an Elman net (see section 3.2.2); previous state ${\bf x}[t-1]$ is called context --in view of the fact that they try to encode information about the inputs seen so far, ${\bf u}[1]\ldots {\bf u}[t-1]$-- and current state ${\bf x}[t]$ is said to be stored in hidden units. Instead of using BPTT or RTRL, the networks are trained using simple backpropagation in each time step without considering the recurrent effects of each weight on the values of context units. Elman studies the performance of this network and the nature of the representations learned by the network when it is trained to perform four sequence prediction tasks: In all cases, Elman (1990) nets learn the temporal structure present in the sequences of events they are trained to predict.


next up previous contents index
Next: #pollack90j###: Up: Papers Previous: #jordan86t###:   Contents   Index
Debian User 2002-01-21