<?xml version="1.0" encoding="UTF-8"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en-gb">
<link rel="self" type="application/atom+xml" href="http://forum.robosavvy.com/feed.php?f=3&amp;t=3430" />

<title>RoboSavvy Forum</title>
<subtitle>Robosavvy Forum: The largest online community of Humanoid Robot Builders</subtitle>
<link href="http://forum.robosavvy.com/index.php" />
<updated>2009-04-06T08:51:32+01:00</updated>

<author><name><![CDATA[RoboSavvy Forum]]></name></author>
<id>http://forum.robosavvy.com/feed.php?f=3&amp;t=3430</id>
<entry>
<author><name><![CDATA[limor]]></name></author>
<updated>2009-04-06T08:51:32+01:00</updated>
<published>2009-04-06T08:51:32+01:00</published>
<id>http://forum.robosavvy.com/viewtopic.php?t=3430&amp;p=19540#p19540</id>
<link href="http://forum.robosavvy.com/viewtopic.php?t=3430&amp;p=19540#p19540"/>
<title type="html"><![CDATA[Computer Program Self-Discovers Laws of Physics]]></title>

<content type="html" xml:base="http://forum.robosavvy.com/viewtopic.php?t=3430&amp;p=19540#p19540"><![CDATA[
thanks for that!!<br />better save a copy before someone decides it is copyrighted and removes the pdf.<p>Statistics: Posted by <a href="http://forum.robosavvy.com/memberlist.php?mode=viewprofile&amp;u=2">limor</a> — Mon Apr 06, 2009 8:51 am</p><hr />
]]></content>
</entry>
<entry>
<author><name><![CDATA[Bullit]]></name></author>
<updated>2009-04-04T05:17:09+01:00</updated>
<published>2009-04-04T05:17:09+01:00</published>
<id>http://forum.robosavvy.com/viewtopic.php?t=3430&amp;p=19531#p19531</id>
<link href="http://forum.robosavvy.com/viewtopic.php?t=3430&amp;p=19531#p19531"/>
<title type="html"><![CDATA[Computer Program Self-Discovers Laws of Physics]]></title>

<content type="html" xml:base="http://forum.robosavvy.com/viewtopic.php?t=3430&amp;p=19531#p19531"><![CDATA[
You mean this science article <img src="http://forum.robosavvy.com/images/smilies/icon_smile.gif" alt=":-)" title="Smile" /><br /><a href="http://ccsl.mae.cornell.edu/sites/default/files/Science09_Schmidt.pdf" class="postlink">http://ccsl.mae.cornell.edu/sites/default/files/Science09_Schmidt.pdf</a><p>Statistics: Posted by <a href="http://forum.robosavvy.com/memberlist.php?mode=viewprofile&amp;u=173">Bullit</a> — Sat Apr 04, 2009 5:17 am</p><hr />
]]></content>
</entry>
<entry>
<author><name><![CDATA[limor]]></name></author>
<updated>2009-04-03T17:51:42+01:00</updated>
<published>2009-04-03T17:51:42+01:00</published>
<id>http://forum.robosavvy.com/viewtopic.php?t=3430&amp;p=19524#p19524</id>
<link href="http://forum.robosavvy.com/viewtopic.php?t=3430&amp;p=19524#p19524"/>
<title type="html"><![CDATA[Computer Program Self-Discovers Laws of Physics]]></title>

<content type="html" xml:base="http://forum.robosavvy.com/viewtopic.php?t=3430&amp;p=19524#p19524"><![CDATA[
I met <a href="http://www.mae.cornell.edu/Lipson/" class="postlink">Hod Lipson</a> last week from Cornell University who talked about an upcoming revolutionary algorithm that can deduct the laws of physics by analyzing data taken from a double-pendulum.<br /><br />He is famed from developing robots that learn to walk by themselves and self-assemble.<br /><br />Hopefully it will be of big benefit to walking robots as it can deduct the laws governing human walking and apply them to humanoids.<br />One thing not mentioned in the <a href="http://blog.wired.com/wiredscience/2009/04/newtonai.html" class="postlink">Wired magazine</a> article (and is probably mentioned in the science magazine which i dont have yet), is that the algorithm has to be able to &quot;change the experiment&quot; in order to succeed in deducting the underlying formula abstraction. This implies in the case of the double pendulum, changing the lengths of the pendulum parts and weights.<br /><br /><a href="http://blog.wired.com/wiredscience/2009/04/newtonai.html" class="postlink"><img src="http://robosavvy.com/Builders/limor/doublependulum.jpg" alt="Image" /></a><p>Statistics: Posted by <a href="http://forum.robosavvy.com/memberlist.php?mode=viewprofile&amp;u=2">limor</a> — Fri Apr 03, 2009 5:51 pm</p><hr />
]]></content>
</entry>
</feed>