<?xml version="1.0"?> 
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "mathml.dtd"> 
<?xml-stylesheet type="text/css" href="thesis.css"?> 
<html  
xmlns="http://www.w3.org/1999/xhtml"  
><head><title> </title> 
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" /> 
<meta name="generator" content="TeX4ht (http://www.cis.ohio-state.edu/~gurari/TeX4ht/mn.html)" /> 
<meta name="originator" content="TeX4ht (http://www.cis.ohio-state.edu/~gurari/TeX4ht/mn.html)" /> 
<!-- 3,early_,early^,xhtml,mozilla --> 
<meta name="src" content="thesis.tex" /> 
<meta name="date" content="2002-08-28 13:56:00" /> 
<link rel="stylesheet" type="text/css" href="thesis.css" /> 
</head><body 
>
 <a 
  name="x34-47001f2"></a>
    In the real SQ model there is no set of examples. The algorithm asks a query <!--l. 1742--><math 
xmlns="http://www.w3.org/1998/Math/MathML" 
mode="inline">
 <mrow 
><mi 
>&#x03C7;</mi></mrow></math> and is given a response <!--l. 1742--><math 
xmlns="http://www.w3.org/1998/Math/MathML" 
mode="inline">
 <mrow 
><msub><mrow 
><mover 
accent="true"><mrow 
><mi 
>D</mi></mrow><mo 
class="MathClass-op">&#x0302;</mo></mover></mrow><mrow 
><mi 
>&#x03C7;</mi></mrow></msub 
></mrow></math> that is guaranteed to be
 near to the true value <!--l. 1742--><math 
xmlns="http://www.w3.org/1998/Math/MathML" 
mode="inline">       <mrow 
><msub><mrow 
><mi 
>D</mi></mrow><mrow 
><mi 
>&#x03C7;</mi></mrow></msub 
></mrow></math>.
 That is, the true SQ model is an abstraction of the scenario described here where <!--l. 1742--><math 
xmlns="http://www.w3.org/1998/Math/MathML" 
mode="inline">
 <mrow 
><msub><mrow 
><mover 
accent="true"><mrow 
><mi 
>D</mi></mrow><mo 
class="MathClass-op">&#x0302;</mo></mover></mrow><mrow 
><mi 
>&#x03C7;</mi></mrow></msub 
></mrow></math> is
 computed from an observed sample.
     
</body> 
</html> 
