# An ode to biological principles

**URL:** https://discourse.numenta.org/t/an-ode-to-biological-principles/10481
**Category:** Lounge
**Created:** [February 1, 2023, 10:37pm UTC](https://discourse.numenta.org/t/an-ode-to-biological-principles/10481 "2023-02-01T22:37:38Z")
**Posts on this page:** 1
**Showing post:** 9

<div class="post-metadata">

### Author: ![bkaz](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/bkaz/32/6455_2.png) [@bkaz](https://discourse.numenta.org/u/bkaz)
#### Post date: [February 3, 2023, 12:05am UTC](https://discourse.numenta.org/t/an-ode-to-biological-principles/10481/9 "2023-02-03T00:05:01Z")

</div>

> [@sebjwallace](#):
>
> The cortex scales it pretty well.

What do you think about dendritic backprop, it seems equivalent to 5-8 layers of conventional backprop? The latest on that:

> [@NN, centroid clustering, search](https://discourse.numenta.org/t/nn-centroid-clustering-search/10443/23):
>
> Ok, here is a single-centroid version of backprop, modeled after dendritic tree:  
> [[2211.11378] Learning on tree architectures outperforms a convolutional feedforward network](https://arxiv.org/abs/2211.11378),  
> [Is Brain Learning Weaker Than Artificial Intelligence? - Neuroscience News](https://neurosciencenews.com/brain-ai-learning-22396/)

BTW, I do agree that greyscale version of Hebbian learning is the core of conventional backprop too, as discussed in the quoted thread.

---

_[View the full topic](https://discourse.numenta.org/t/an-ode-to-biological-principles/10481)._