Next: #tino95j###
Up: Inference of finite-state machines
Previous: #NC:Giles92###:
  Contents
  Index
use Elman's
first-order architecture
((Elman, 1990)), as
Cleeremans et al. (1989) , but
train it to classify strings in small learning sets as grammatical or
ungrammatical --according to four of
the grammars in Tomita (1982)
-- using
backpropagation through time (Rumelhart et al., 1986) with batch
updating (see section 3.4.1). Once the
DTRNN has been trained, automata are extracted using a special
clustering algorithm
that has been
described in section 5.4. One of the most interesting
features of the paper is its graphical study of the learning process,
and , in particular, that of the size of clusters as training
progresses. The size of the clusters is proposed as an indicator of
generalization ability.
Debian User
2002-01-21