<?xml version="1.0" encoding="utf-8" ?>

<rss version="2.0" 
   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
   xmlns:admin="http://webns.net/mvcb/"
   xmlns:dc="http://purl.org/dc/elements/1.1/"
   xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
   xmlns:wfw="http://wellformedweb.org/CommentAPI/"
   xmlns:content="http://purl.org/rss/1.0/modules/content/"
   >
<channel>
    
    <title>Loligo Blog - Science &amp; Tech</title>
    <link>https://www.standoutpublishing.com/Blog/</link>
    <description>Neural Networks &amp; Robotics</description>
    <dc:language>en</dc:language>
    <generator>Serendipity 2.1.6 - http://www.s9y.org/</generator>
    <pubDate>Mon, 12 Oct 2026 18:08:03 GMT</pubDate>

    <image>
    <url>http://standoutpublishing.com/Res/Image/NN137x150Wht.jpg</url>
    <title>RSS: Loligo Blog - Science &amp; Tech - Neural Networks &amp; Robotics</title>
    <link>https://www.standoutpublishing.com/Blog/</link>
    <width>137</width>
    <height>150</height>
</image>

<item>
    <title>Introducing: Influence Learning</title>
    <link>https://www.standoutpublishing.com/Blog/archives/53-Introducing-Influence-Learning.html</link>
            <category>Neural Networks</category>
            <category>Pub notes</category>
            <category>Science &amp; Tech</category>
    
    <comments>https://www.standoutpublishing.com/Blog/archives/53-Introducing-Influence-Learning.html#comments</comments>
    <wfw:comment>https://www.standoutpublishing.com/Blog/wfwcomment.php?cid=53</wfw:comment>

    <slash:comments>0</slash:comments>
    <wfw:commentRss>https://www.standoutpublishing.com/Blog/rss.php?version=2.0&amp;type=comments&amp;cid=53</wfw:commentRss>
    

    <author>nospam@example.com (John R)</author>
    <content:encoded>
    &lt;br /&gt;&lt;br /&gt;
&lt;br /&gt;
&lt;span style=&quot;font-style:italic&quot;&gt;&lt;span style=&quot;font-weight:bold&quot;&gt;&lt;a href=&quot;http://standoutpublishing.com/g/Influence-Learning.html&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;Influence learning&lt;/a&gt;&lt;/span&gt;&lt;/span&gt; is one of two new learning algorithms that have emerged (so far) from the Netlab development effort. This blog entry contains a brief overview describing how it works, and some of the advantages it brings to the task of neural network weight-adjustment.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;h2 class=&quot;SecHeader&quot;&gt;&lt;br /&gt;
How It Works&lt;br /&gt;
&lt;/h2&gt;&lt;br /&gt;
&lt;br /&gt;
This learning method is based on the notion that&amp;mdash;like their collective counterparts&amp;mdash;neurons may be attracted to, and occasionally repulsed by, the exercise of influence by others. In the case of neurons, the &quot;others&quot; would be other neurons.  As simple as that notion sounds, it produces a learning method with a number of interesting benefits and advantages over the current crop of learning algorithms.&lt;br /&gt;
&lt;br /&gt;
A neuron using &lt;span style=&quot;font-style:italic&quot;&gt;&lt;span style=&quot;font-weight:bold&quot;&gt;&lt;a href=&quot;http://standoutpublishing.com/g/Influence-Learning.html&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;influence learning&lt;/a&gt;&lt;/span&gt;&lt;/span&gt; is not nosy, and does not concern itself with &lt;span style=&quot;font-weight:bold&quot;&gt;&lt;span style=&quot;font-style:italic&quot;&gt;how&lt;/span&gt;&lt;/span&gt; its post-synaptic (forward) neurons are learning. It simply trusts that their job is to learn, and that they are doing their job. In other words, a given neuron fully expects, and assumes that other neurons within the system are learning. Each one treats post-synaptic neurons that are exercising the most influence as role models for adjusting connection-strengths. The norm is for neurons to see influential forward neurons as positive role models, but neurons may also see influential forward neurons as negative role models.&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;h2 class=&quot;SecHeader&quot;&gt;&lt;br /&gt;
It Is Simple&lt;br /&gt;
&lt;/h2&gt;&lt;br /&gt;
&lt;br /&gt;
As you might guess, the first benefit is simplicity. The method does not try to hide a lack of new ideas behind a wall of new computational complexity. It is a simple, new, method based on a simple, almost axiomatic, observation, and it can be implemented with relatively little computational power. Diagrammatically: &lt;br /&gt;
&lt;br /&gt;
&lt;center&gt;&lt;br /&gt;
&lt;img src=&quot;https://standoutpublishing.com/Site/gloss/oo/Image/InfluenceLearningTargetNeuronDiag.jpg&quot; alt=&quot;Diagram showing how influence is garnered from post-synaptic neurons and the effect it has on pre-synaptic neuons&quot; width=&quot;75%&quot; border=&quot;0&quot;&gt;&lt;br /&gt;
&lt;/center&gt;&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;h2 class=&quot;SecHeader&quot;&gt;&lt;br /&gt;
It Imposes No Restrictions On Feedback&lt;br /&gt;
&lt;/h2&gt;&lt;br /&gt;
&lt;br /&gt;
&lt;span style=&quot;font-style:italic&quot;&gt;&lt;span style=&quot;font-weight:bold&quot;&gt;&lt;a href=&quot;http://standoutpublishing.com/g/Influence-Learning.html&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;Influence Learning&lt;/a&gt;&lt;/span&gt;&lt;/span&gt; is completely free of feedback restrictions. That is, network connection-structures may be designed with any type, or amount of feedback looping. The learning mechanism will continue to be able to properly adapt connection-strengths regardless of how complex the feedback scheme is. The types of feedback designers are free to employ include servo feedback, which places the outside world (or some network structure that is closer to the outside world) directly in the signaling feedback path.&lt;br /&gt;
&lt;br /&gt;
This type of &quot;servo-feedback&quot; is shown graphically in figure 6-5 of the book, which has been re-produced here.&lt;br /&gt;
&lt;center&gt;&lt;br /&gt;
&lt;img src=&quot;http://standoutpublishing.com/Res/Blog/BookFig6-5_450x500.gif&quot; border=&quot;0&quot;&gt;&lt;br /&gt;
&lt;/center&gt;&lt;br /&gt;
&lt;br /&gt;
 &lt;a class=&quot;block_level&quot; href=&quot;https://www.standoutpublishing.com/Blog/archives/53-Introducing-Influence-Learning.html#extended&quot;&gt;Continue reading &quot;Introducing: Influence Learning&quot;&lt;/a&gt;
    </content:encoded>

    <pubDate>Tue, 07 Sep 2010 16:14:00 +0000</pubDate>
    <guid isPermaLink="false">https://www.standoutpublishing.com/Blog/archives/53-guid.html</guid>
    <category>cognition-perception</category>
<category>Influence-Learning</category>
<category>Memory</category>
<category>Netlab</category>
<category>Neural-Networks</category>
<category>Patents</category>

</item>
<item>
    <title>Influence Learning Gets A Patent</title>
    <link>https://www.standoutpublishing.com/Blog/archives/58-Influence-Learning-Gets-A-Patent.html</link>
            <category>Announcements</category>
            <category>Neural Networks</category>
            <category>Science &amp; Tech</category>
    
    <comments>https://www.standoutpublishing.com/Blog/archives/58-Influence-Learning-Gets-A-Patent.html#comments</comments>
    <wfw:comment>https://www.standoutpublishing.com/Blog/wfwcomment.php?cid=58</wfw:comment>

    <slash:comments>0</slash:comments>
    <wfw:commentRss>https://www.standoutpublishing.com/Blog/rss.php?version=2.0&amp;type=comments&amp;cid=58</wfw:commentRss>
    

    <author>nospam@example.com (John R)</author>
    <content:encoded>
    &lt;br /&gt;&lt;br /&gt;
&lt;table width=&quot;100%&quot;&gt;
&lt;tr&gt;
&lt;td width=&quot;35%&quot; valign=&quot;top&quot;&gt;
&lt;img src=&quot;http://standoutpublishing.com/Site/ooRes/Blog/USPTOSealSmaller.jpg&quot;&gt;
&lt;/td&gt;
&lt;td width=&quot;1%&quot;&gt;
    &amp;#160;
&lt;/td&gt;
&lt;td width=&quot;62%&quot; valign=&quot;top&quot;&gt;
&lt;h2 class=&quot;SecHeader&quot;&gt;
Woo HOOO!
&lt;/h2&gt;
&lt;br /&gt;&lt;br /&gt;

&lt;a href=&quot;http://standoutpublishing.com/Blog/archives/53-Introducing-Influence-Learning.html&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;Influence Based Learning&lt;/a&gt;, one of two new learning methods described in the book &lt;a href=&quot;http://standoutpublishing.com/Prod/Book/Netlabv03a/&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;Netlab Loligo&lt;/a&gt;, has just been awarded a United States Patent. The official title of the patent is:
&lt;br /&gt;&lt;br /&gt;

&lt;center&gt;
&lt;span style=&quot;font-weight:bold&quot;&gt;&amp;ldquo;Feedback-Tolerant Method And Device Producing Weight-Adjustment Factors For Pre-Synaptic Neurons In Artificial Neural Networks&amp;rdquo;&lt;/span&gt;
&lt;/center&gt;
&lt;br /&gt;

The title is a mouthful, primarily designed to help future patent searchers determine if their great idea has already been discovered and patented. It is fully described and discussed in the book, where it is simply referred to as &lt;a href=&quot;http://standoutpublishing.com/Blog/archives/53-Introducing-Influence-Learning.html&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;Influence Learning&lt;/a&gt;.
&lt;br /&gt;&lt;br /&gt;

As the patent-title expresses, one of the benefits it imparts over existing learning algorithms, is that it is feedback-tolerant. It will work fine with the current-day feed-forward networks configured as &quot;slabs&quot;, but it also allows connecting neurons to pre-synaptic neurons as well. That is, it allows feedback, which means you don&#039;t have to configure your network with &quot;&lt;a href=&quot;http://standoutpublishing.com/g/Hidden-Layer.html&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;hidden layers&lt;/a&gt;&quot; anymore if you don&#039;t want to. You are free to use &lt;a href=&quot;http://standoutpublishing.com/Blog/archives/52-Brain-Wiring-Structure-How-About-a-Donut.html&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;any connectome you&#039;d like&lt;/a&gt;.
&lt;/td&gt;
&lt;td width=&quot;2%&quot;&gt;
    &amp;#160;
&lt;/td&gt;
&lt;/tr&gt;
&lt;/table&gt;&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;ul&gt;

&lt;li&gt; &lt;a href=&quot;https://standoutpublishing.com/Blog/archives/53-Introducing-Influence-Learning.html&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;Introducing: Influence Learning&lt;/a&gt;&lt;p /&gt;&lt;/li&gt;

&lt;li&gt; &lt;a href=&quot;http://standoutpublishing.com/Doc/o/Patents/07814038/07814038-01.pdf&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;[pdf] Patent Title Page&lt;/a&gt;&lt;p /&gt;&lt;/li&gt;
&lt;!--
&lt;li&gt; &lt;a href=&quot;http://www.uspto.gov/web/patents/patog/week41/OG/html/1359-2/US07814038-20101012.html&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;Patent Office - Official Gazette Entry &lt;span style=&quot;font-weight:bold&quot;&gt;[TEMPORARY LINK]&lt;/span&gt;&lt;/a&gt;&lt;p /&gt;&lt;/li&gt;
--&gt;
      &lt;li&gt;&lt;a href=&quot;http://standoutpublishing.com/Blog/archives/57-Neural-Networks-Backgrounder-Ce-nest-pas-une-lhistoire.html&quot; target=&quot;_blank&quot; class=&quot;bb-url&quot;&gt;Neural Networks Backgrounder: Ce n&#039;est pas une l&#039;histoire&lt;/a&gt;&lt;br /&gt;
        A quick backgrounder on neural networks presented in a sketchy semi-historical format.
          &lt;/li&gt;

&lt;/ul&gt;&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
&lt;br /&gt;
 
    </content:encoded>

    <pubDate>Sun, 17 Oct 2010 13:31:00 +0000</pubDate>
    <guid isPermaLink="false">https://www.standoutpublishing.com/Blog/archives/58-guid.html</guid>
    <category>Influence-Learning</category>
<category>Netlab</category>
<category>Neural-Networks</category>
<category>Patents</category>

</item>

</channel>
</rss>
