<?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>1.1 The learning problem</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 
>
   <div class="crosslinks"><p class="noindent">[<a 
href="thesisse2.xml" >next</a>] [<a 
href="#tailthesisse1.xml">tail</a>] [<a 
href="thesisch1.xml#thesisse1.xml" >up</a>] </p></div>
   <h3 class="sectionHead"><span class="titlemark">1.1. </span> <a 
  name="x5-40001.1"></a>The learning problem</h3>
<!--l. 113--><p class="noindent">What is learning? Learning is the process of discovering relationships between
events. Knowledge of the relationships between events is of great importance in
coping with the environment. Naturally, humans tend to be quite good at the
process of learning&#x2014;so good that we sometimes do not even realize when it is
hard.
</p><!--l. 119--><p class="indent">   The learning problem which we focus on is learning for computers. In particular,
&#x201C;How can a computer learn?&#x201D; A few examples are illustrative:
</p><!--l. 122--><p class="indent">
           </p><ol type="1" class="enumerate1" start="1" 
>
        <li class="enumerate"><a 
  name="x5-4002x1"></a>How can we make a computer with a microphone output a text version of
        what is being spoken?
           </li>
        <li class="enumerate"><a 
  name="x5-4004x2"></a>How can we make a computer with a camera recognize John?
           </li>
        <li class="enumerate"><a 
  name="x5-4006x3"></a>How can we make a computer controlling a robot arrive at some location?</li></ol>
<!--l. 127--><p class="nopar"> We will work on learning a function from some input space to some output space - the
                                                                     

                                                                     
supervised learning model.
</p><!--l. 132--><p class="indent">
                                                                     

                                                                     
</p>
   <div class="crosslinks"><p class="noindent">[<a 
href="thesisse2.xml" >next</a>] [<a 
href="thesisse1.xml" >front</a>] [<a 
href="thesisch1.xml#thesisse1.xml" >up</a>] </p></div><a 
  name="tailthesisse1.xml"></a>  
</body> 
</html> 
