<?xml version="1.0" standalone="yes"?>
<Paper uid="N01-1028">
  <Title>Learning optimal dialogue management rules by using reinforcement learning and inductive logic programming</Title>
  <Section position="7" start_page="0" end_page="0" type="evalu">
    <SectionTitle>
5 Discussion
</SectionTitle>
    <Paragraph position="0"> Recent work on reinforcement learning and dialogue management has mainly focused on how to reduce the search space for the optimal strategy. Because reinforcement learning is state based and there may potentially be a large number of states, problems may arise when few dialogues are available and the data too sparse to select the best strategy. States can usually be collapsed to make this problem less acute. The main idea here is to express the state of the dialogue by a limited number of features while keeping enough and the right kind of information to be able to learn useful strategies (Walker et al., 1998). There has also been new research on how to model the dialogue with partially observable Markov models (Roy et al., 2000).</Paragraph>
    <Paragraph position="1"> Some work has also been done on nding out rules to select dialogue management strategies.</Paragraph>
    <Paragraph position="2"> For example, Litman and Pan (2000) use machine learning to learn rules detecting when dialogues go badly. The dialogue manager uses a strategy prede ned by a dialogue designer.</Paragraph>
    <Paragraph position="3"> If a rule detects a bad dialogue, the dialogue strategy is changed to a more restrictive, more system guided strategy. Our approach is different from that work since the strategy is not prede ned but based on the optimal strategy found by reinforcement learning. Our rules not only detect in principle when a dialogue is going badly but also indicate which action to take.</Paragraph>
    <Paragraph position="4"> The e ciency of the rules obviously depends on the way the optimal strategy search space has been modeled and other conditions in uencing learning.</Paragraph>
    <Paragraph position="5"> Some pieces of work have been concerned with natural language processing from an inductive logic programming point of view. Notably, work on morphology (Mooney and Cali , 1995) and parsing (Thompson et al., 1997) has been carried out. However, as far as we know, the application of inductive logic programming to dialogue management is new.</Paragraph>
  </Section>
class="xml-element"></Paper>