| | |
Summary: In Proc. of the International Conference on Neural Information Processing (ICONIP'96), Hong Kong, 1996
Neural Networks and Child Language Development:
Towards a `Conglomerate' Neural Network Simulation Architecture
Syed Sibte Raza Abidi
School of Computer Sciences
Universiti Sains Malaysia, Penang, MALAYSIA
E-Mail: sraza@cs.usm.my
Abstract
Neural networks provide a basis for studying child language development in that such networks emphasise
learning. We report a simulation of some key aspects of child language development during infancy. We
argue that in order to simulate the uniquely human language learning, it is important to use a `conglomerate'
neural network architecture that integrates the collective strengths of a variety of neural networks in some
principled fashion to take into account the diverse nature of inputs to and outputs from a child learning
language. We present such a `conglomerate' neural network architecture - ACCLAIM that integrates both
supervised and unsupervised learning algorithms, to simulate the learning of concepts, words, conceptual and
semantic relations and simple word-order rules, thus mimicking the production of child-like one-word and two-
word language. The simulations carried out are `language informed' as realistic child language data has been
used for training the neural networks.
1. Introduction
Neural network community is keenly interested in simulating human learning, and indeed language learning
|
