# Intelligence is embarrassingly simple

**URL:** <https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930>\
**Category:** General Neuroscience\
**Created:** [August 11, 2023, 10:03pm UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930 "2023-08-11T22:03:46Z")\
**Posts on this page:** 12\
**Page:** 2

<div class="post-metadata">

**Author:** ![Bitking](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/bitking/32/3099_2.png) [@Bitking](https://discourse.numenta.org/u/Bitking)\
**Post date:** [August 17, 2023, 7:00pm UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/22 "2023-08-17T19:00:20Z")

</div>

> [@Bullbash](#):
>
> a) conventional-systemic studying “behavior of cells” and then understanding the neural code (carbon to silicon)

From my point of view, I see that you are more into the same general approach as BKAZ. Good luck in your endeavors.

I think that much of the brain function/computation is buried in the configuration of the sub-components. I see that this configuration is based around the spatial/temporal behavior of the components.

As long as the human brain is the only real example of human level intelligence, I will continue to look there for inspiration and understading in building an AI.

---

<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:** [August 17, 2023, 7:49pm UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/23 "2023-08-17T19:49:04Z")

</div>

> [@Bitking](#):
>
> From my point of view, I see that you are more into the same general approach as BKAZ.

I see nothing in common.

---

<div class="post-metadata">

**Author:** ![Bitking](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/bitking/32/3099_2.png) [@Bitking](https://discourse.numenta.org/u/Bitking)\
**Post date:** [August 17, 2023, 7:53pm UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/24 "2023-08-17T19:53:04Z")

</div>

You are both chasing an algorithm rather than slavishly copying the brain.  
Yes, different algorithms, but each is convinced that you are pursuing the holy grail computation.

---

<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:** [August 17, 2023, 8:26pm UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/25 "2023-08-17T20:26:16Z")

</div>

That’s like saying a space rock and Andromeda nebula are similar because neither is on Earth. A very narrow-minded POV, if I may. One similarity is instance-based learning, but that depends on how you define instances.

---

<div class="post-metadata">

**Author:** ![Bitking](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/bitking/32/3099_2.png) [@Bitking](https://discourse.numenta.org/u/Bitking)\
**Post date:** [August 17, 2023, 8:53pm UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/26 "2023-08-17T20:53:53Z")

</div>

And each thinks they have the one true answer.

---

<div class="post-metadata">

**Author:** ![Bullbash](https://avatars.discourse-cdn.com/v4/letter/b/e68b1a/32.png) [@Bullbash](https://discourse.numenta.org/u/Bullbash)\
**Post date:** [August 17, 2023, 8:55pm UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/27 "2023-08-17T20:55:01Z")

</div>

The perfect conclusion to the perfect discussion:

Afoot and light-hearted I take to the open road,  
Healthy, free, the world before me,  
The long brown path before me leading wherever I choose.

Henceforth I ask not good-fortune, I myself am good-fortune,  
Henceforth I whimper no more, postpone no more, need nothing,  
Done with indoor complaints, libraries, querulous criticisms,

Strong and content I travel the open road.(c)

---

<div class="post-metadata">

**Author:** ![cezar\_t](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/cezar_t/32/4451_2.png) [@cezar\_t](https://discourse.numenta.org/u/cezar_t)\
**Post date:** [August 18, 2023, 4:57am UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/28 "2023-08-18T04:57:43Z")

</div>

> [@Bullbash](#):
>
> I was convinced the neuron’s owner is the brain owner (intellectual entity?), which tend to adapt and survive.

If you dont have time to explain it, a link to some example code where those analogue inputs are used would be more useful than a slightly ironic non-answer.

---

<div class="post-metadata">

**Author:** ![neel\_g](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/neel_g/32/5503_2.png) [@neel\_g](https://discourse.numenta.org/u/neel_g)\
**Post date:** [August 18, 2023, 9:48pm UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/29 "2023-08-18T21:48:51Z")

</div>

> [@Bullbash](#):
>
> The main point though, there are interesting symbolic operations over “named nodes-patterns” , “faceless” will stay pure stochastic forever. A shortcut to a dead-end. BP is not local.

I wouldn’t be too bearish; you wouldn’t believe how many alternative methods towards AGI I’ve heard in my short lifetime. But the bitter lesson always comes knocking at the door - and those who ignore it are bound to fail 🙂

BP+NNs works much better than most people credit them for; and many of their failings are statistical limits rather than being an intrinsic drawback of them (Indeed, lacking inductive biases is the sole reason they outperform other methods so well in the first place).

Its a dangerous game, where being seduced by biology and ignoring the theory could easily lead to a dead-end…

---

<div class="post-metadata">

**Author:** ![complyue](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/complyue/32/6056_2.png) [@complyue](https://discourse.numenta.org/u/complyue)\
**Post date:** [August 22, 2023, 3:55pm UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/30 "2023-08-22T15:55:09Z")

</div>

In reading [Myths of the Instance Based Learning](https://www.linkedin.com/pulse/myths-instance-based-learning-alex-semenov) I still can’t figure out what’s the architecture of the ANN, especially failed finding transcript/explanation of this video:

[![](https://img.youtube.com/vi/uAQB9d6ovlk/maxresdefault.jpg "USL, SL & RL Hebbian online learner") ](https://www.youtube.com/watch?v=uAQB9d6ovlk)

I’m a hardcore software engineering guy, yet failed cracking [GitHub - MasterAlgo/GPT-Teaser](https://github.com/MasterAlgo/GPT-Teaser) for lacking comments there.

Also many important (to my understanding, at least) LinkedIn links are not available now, e.g. in this page:

> **[Alex Semenov on LinkedIn: Neuromorphic ANN growing billions of connections....](https://www.linkedin.com/posts/activity-6873695912426917889-Rjri/)**
>
> Neuromorphic ANN growing billions of connections. Companion to https://lnkd.in/gH\_YVgg9
> Beyond learning online (88% accuracy on first 10% of IMDB dataset) it…

```auto
Math:
https://lnkd.in/gnyv-r-y
https://lnkd.in/g--EERUk
Generations (both word and character levels):
https://lnkd.in/gN8xC-mt
https://lnkd.in/gGKMd-hF
https://lnkd.in/gJdmNJf3
https://lnkd.in/g-cFSa8w

```

Hell, I want to understand the design better, any help appreciated!

---

<div class="post-metadata">

**Author:** ![complyue](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/complyue/32/6056_2.png) [@complyue](https://discourse.numenta.org/u/complyue)\
**Post date:** [August 23, 2023, 9:38am UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/31 "2023-08-23T09:38:36Z")

</div>

@Bullbash I especially have difficulty in understanding this:

[Myths of the Instance Based Learning](https://www.linkedin.com/pulse/myths-instance-based-learning-alex-semenov/#:~:text=Systems%20(instances)%20similarity%20depends%20on%20number%20of%20common%20subsystem%20(tokens)).

> This net generalizes well by the rule:
> 
> Systems (instances) similarity depends on number of common subsystem (tokens).
> 
> Some distance measures behave better then others but in general a successful comparison is made on sets of millions common constituents. That is like comparing real-component vectors in 10^6 dimensional space. It just works, it does. The Curse of Dimensionality?! Hello?
> 
> The intuition of such a network is in this crude visualization:
> 
> [https://youtu.be/uAQB9d6ovlk](https://youtu.be/uAQB9d6ovlk)

In your schema, does a system/instance have fixed number of subsystems, or the number varies? Also does the order of subsystems matter? I think I’m asking: how the identity of a subsystem/token/system/instance form?

* * *

Do you mean “comparing real-component vectors in 10^6 dimensional space” doesn’t suffer the Curse of Dimensionality? AFAICT Euclidean distance between such vectors doesn’t work sufficiently well, do you imply that some distances are inherently free of Dim Curse? What are those?

---

<div class="post-metadata">

**Author:** ![Bullbash](https://avatars.discourse-cdn.com/v4/letter/b/e68b1a/32.png) [@Bullbash](https://discourse.numenta.org/u/Bullbash)\
**Post date:** [August 24, 2023, 1:04am UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/32 "2023-08-24T01:04:08Z")

</div>

> [@neel\_g](#):
>
> being seduced by biology

should you tell it to Jeff Hawkins? 🙂 ))

> [@neel\_g](#):
>
> those who ignore it are bound to fail 🙂

will keep it in mind…  
Thank you for contributing!

---

<div class="post-metadata">

**Author:** ![neel\_g](https://yyz2.discourse-cdn.com/flex030/user_avatar/discourse.numenta.org/neel_g/32/5503_2.png) [@neel\_g](https://discourse.numenta.org/u/neel_g)\
**Post date:** [August 24, 2023, 9:58am UTC](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930/33 "2023-08-24T09:58:17Z")

</div>

> [@Bullbash](#):
>
> should you tell it to Jeff Hawkins? 🙂 ))

Ironic, considering Numenta’s current direction has been leaning more and more heavily into integrating their theories with DL. Hawkins’ is smart enough to realize where this theories are flawed. And smart enough to lean onto other concurrent work to plug holes in his own theories. Hopefully, he might be able to make some discoveries in parallel to both the fields but only time will tell…

[Previous page](https://discourse.numenta.org/t/intelligence-is-embarrassingly-simple/10930.md?page=1)