<?xml version="1.0" standalone="yes"?>
<Paper uid="P91-1018">
  <Title>regier@cogsci.Berkeley.ED U * TR &amp;quot;Above&amp;quot; Figure 1: Learning to Associate Scenes with Spatial Terms</Title>
  <Section position="8" start_page="144" end_page="144" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> The system presented here learns perceptually-grounded semantics for the core senses of eight English prepositions, successfully generalizing to scenes involving landmarks to which the system had not been previously exposed. Moreover, the principle of mutual exclusivity is successfully used to allow learning without explicit negative instances, despite the false negatives in the resulting training sets.</Paragraph>
    <Paragraph position="1"> Current research is directed at extending this work to the case of arbitrarily shaped trajectors, and to handling polysemy. Work is also being directed toward the learning of non-English spatial systems.</Paragraph>
  </Section>
class="xml-element"></Paper>