<node id="583842">
  <nid>583842</nid>
  <type>external_news</type>
  <uid>
    <user id="28466"><![CDATA[28466]]></user>
  </uid>
  <created>1479149085</created>
  <changed>1479149085</changed>
  <title><![CDATA[Video games where people matter? The strange future of emotional AI]]></title>
  <body><![CDATA[<p>The Guardian talks about&nbsp;Mark Riedl, an&nbsp;associate professor at Georgia Tech and director of the university&rsquo;s Entertainment Intelligence Lab. He&rsquo;s currently working on a project named Quixote, an AI system designed to make it easier for non-specialist programmers to create intelligent &lsquo;virtual agents&rsquo;. Quixote allows AI agents to learn social rules and behaviours through reading stories sourced online; right now Riedl uses Amazon&rsquo;s online crowdsourcing marketplace Mechanical Turk to commission people to write short stories around set themes. To demonstrate the concept, Riedl developed a game called Robbery World in which an AI agent has to rob a bank: it studies a series of bank robbery stories submitted on Mechanical Turk; learns the common elements (travel to bank, go to counter, pull out gun, demand cash) and is then rewarded for carrying out any action that advances the plot.</p>

<p>Read the rest of the article here:&nbsp;<a href="https://www.theguardian.com/technology/2016/oct/12/video-game-characters-emotional-ai-developers">https://www.theguardian.com/technology/2016/oct/12/video-game-characters-emotional-ai-developers</a></p>
]]></body>
  <field_article_url>
    <item>
      <url><![CDATA[https://www.theguardian.com/technology/2016/oct/12/video-game-characters-emotional-ai-developers]]></url>
      <title><![CDATA[]]></title>
    </item>
  </field_article_url>
  <field_publication>
    <item>
      <value><![CDATA[ Christine Angelini ]]></value>
    </item>
  </field_publication>
  <field_dateline>
    <item>
      <value>2016-10-12</value>
      <timezone></timezone>
    </item>
  </field_dateline>
  <field_media>
        </field_media>
  <og_groups>
          <item>1299</item>
      </og_groups>
  <og_groups_both>
          <item><![CDATA[GVU Center]]></item>
      </og_groups_both>
    <field_userdata>
      <![CDATA[]]>
  </field_userdata>
</node>
