# Self structuring trees with vector pattern recognisers

**URL:** <https://discourse.numenta.org/t/self-structuring-trees-with-vector-pattern-recognisers/736>\
**Category:** Lounge\
**Created:** [June 9, 2016, 4:17am UTC](https://discourse.numenta.org/t/self-structuring-trees-with-vector-pattern-recognisers/736 "2016-06-09T04:17:05Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![Fraz\_J](https://avatars.discourse-cdn.com/v4/letter/f/85f322/32.png) [@Fraz\_J](https://discourse.numenta.org/u/Fraz_J)\
**Post date:** [June 9, 2016, 4:17am UTC](https://discourse.numenta.org/t/self-structuring-trees-with-vector-pattern-recognisers/736/1 "2016-06-09T04:17:05Z")

</div>

You have some vector pattern recogniser that can respond to say a few hundred input patterns. When it sees a known context in the input vector (I) it outputs a vector (A). Both that output (A) and the input data (I) are fed to another pattern recogniser (output B). For a particular value of A, the value of B can be decided depending on some other context information in the input. This is tree like splitting. You can keep going with another pattern recogniser that takes I and B as input.  
That should work with the random projection neural nets I have and also with Numenta’s neural pattern recognisers as well. In my case I understand how I should go about training the system, I’m not too sure how you would do that with Numenta neurons.

---

<div class="post-metadata">

**Author:** ![floybix](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/floybix/32/46_2.png) [@floybix](https://discourse.numenta.org/u/floybix)\
**Post date:** [June 11, 2016, 3:45am UTC](https://discourse.numenta.org/t/self-structuring-trees-with-vector-pattern-recognisers/736/2 "2016-06-11T03:45:47Z")

</div>

Could you describe the problem you are trying to solve with this?

---

<div class="post-metadata">

**Author:** ![Fraz\_J](https://avatars.discourse-cdn.com/v4/letter/f/85f322/32.png) [@Fraz\_J](https://discourse.numenta.org/u/Fraz_J)\
**Post date:** [June 11, 2016, 2:23pm UTC](https://discourse.numenta.org/t/self-structuring-trees-with-vector-pattern-recognisers/736/3 "2016-06-11T14:23:14Z")

</div>

Data driven neural nets without back-propagation. The idea is to focus learning (more tree branches) in parts of the problem state space where that is needed . I’ll let you know how it turns out in a few days. I’m going to try with a recurrent version I thought of.  
There are some disparate alternatives to back propagation scattered around in research papers. It would be worthwhile to collate them, if anyone ever had the time.

---

<div class="post-metadata">

**Author:** ![Fraz\_J](https://avatars.discourse-cdn.com/v4/letter/f/85f322/32.png) [@Fraz\_J](https://discourse.numenta.org/u/Fraz_J)\
**Post date:** [June 11, 2016, 3:05pm UTC](https://discourse.numenta.org/t/self-structuring-trees-with-vector-pattern-recognisers/736/4 "2016-06-11T15:05:18Z")

</div>

Eg: [http://www.itp.uni-bremen.de/complex/pdf/prl003013.pdf](http://www.itp.uni-bremen.de/complex/pdf/prl003013.pdf)

---

<div class="post-metadata">

**Author:** ![floybix](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/floybix/32/46_2.png) [@floybix](https://discourse.numenta.org/u/floybix)\
**Post date:** [June 12, 2016, 6:53am UTC](https://discourse.numenta.org/t/self-structuring-trees-with-vector-pattern-recognisers/736/5 "2016-06-12T06:53:22Z")

</div>

Interesting. Sounds like Growing Neural Gas, which I know some people here are working with.