<?xml version="1.0" standalone="yes"?>
<Paper uid="C94-1034">
  <Title>MODULARITY IN A CONNECTIONIST MODEL MORPHOLOGY ACQUISITION</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
INTRODUCTION
</SectionTitle>
    <Paragraph position="0"> For many natural languages, the complexity of bound morphology makes it a potentially challenging problem for a learning system, wl, ether human or machine. A language learner must acquire both the ability to map polymorphemlc words onto the sets of semantic elements they tel)resent and to map meanings onto polymorphemic words.</Paragraph>
    <Paragraph position="1"> Unlike previous work on connection,st morphology (e.g., MacWhinney ~5 Leinbaeh (1991), Plunker, &amp; Marehman (1991) and Rumelhart &amp; MeClelland (1986)), the focus of this paper is receptive nmrphology, which represents the more fundamental, or at least the earlier, process, one which productive nmrphology presumably buihls on.</Paragraph>
    <Paragraph position="2"> The task of learning receptive morphology is viewed here ,as follows. The learner is &amp;quot;trained&amp;quot; on pairs of forms, consisting of sequcnces of phones, and &amp;quot;meanings&amp;quot;, consisting of sets of roots and inflections. I will refer to the task as root and inflection identification. Generalization is tested by presenting the learner with words consisting of novel combinations of familiar morphemes. If the rule in question has been acquired, the learner is able to identify the root and inflections in the test word.</Paragraph>
    <Paragraph position="3"> Of interest is whether a model is capable of acquiring rules of all of the types known for natural languages. This paper describes a psychologically motivated connection,st model (Modular Connection,st Network for the Acquisition of Morphology, MCNAM) which approaches this level of performance. The emphasis here is on the role of modularity at the level of root and inflection in the model. I show how this sort of modularity improves performance (lramatically and consider how a network might learn to use modules it is provided with.</Paragraph>
    <Paragraph position="4"> A sel)arate paper (Gasser, 1994) looks in detail at the model's performance for particular categories of morI)hology, in particular, template morphology and reduplication.</Paragraph>
    <Paragraph position="5"> The paper is organized as folh)ws. I first provide a brief overview of the categories of morphological rules found in the werhl's languages. I then present a simple version of the model and discuss simulations which demonstrate that it generalizes for most kinds of morphoh)gical rules. I then describe a version of the model augmented with modularity at the level of root and inflection which generalizes significantly better and show why this appears to be the case. Finally, I describe some tentative attempts to develop a model which is provided with modules and learns how to use them to solve the morphology identification tasks it is faced with.</Paragraph>
  </Section>
class="xml-element"></Paper>