# Some Aspects of Backpropagation

**URL:** <https://discourse.numenta.org/t/some-aspects-of-backpropagation/11709>\
**Category:** Lounge\
**Created:** [February 3, 2025, 5:26am UTC](https://discourse.numenta.org/t/some-aspects-of-backpropagation/11709 "2025-02-03T05:26:05Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![SeanOConnor](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/seanoconnor/32/3441_2.png) [@SeanOConnor](https://discourse.numenta.org/u/SeanOConnor)\
**Post date:** [February 3, 2025, 5:26am UTC](https://discourse.numenta.org/t/some-aspects-of-backpropagation/11709/1 "2025-02-03T05:26:05Z")

</div>

Understanding backpropagation as a linear update with deferred non-linear effects:  
[https://sites.google.com/view/algorithmshortcuts/some-aspects-of-backpropagation](https://sites.google.com/view/algorithmshortcuts/some-aspects-of-backpropagation)

The thing I learned from writing about it is if you want to use locality sensitive hashing to select weights, weight vectors or weight matrices from a pool you better adopt some tactic to mitigate the over-response to minor Gaussian noise changes.  
Possibly by some kind of blending technique.  
I will go away and ruminate about it.

---

<div class="post-metadata">

**Author:** ![SeanOConnor](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/seanoconnor/32/3441_2.png) [@SeanOConnor](https://discourse.numenta.org/u/SeanOConnor)\
**Post date:** [February 4, 2025, 5:39am UTC](https://discourse.numenta.org/t/some-aspects-of-backpropagation/11709/2 "2025-02-04T05:39:25Z")

</div>

It seems like you should use locality sensitive hashing to select individual weights from a pool of weights, rather than complete weight vectors or matrices. That seems less likely to disrupt backpropagation with excessive non-linear behavior. Getting one weight wrong in a n dimensional weighted sum is not especially disruptive, for example.