[ExI] Neuromorphic Chips

[Brent Allsop]

Very interesting.



Stuart asks Colin:



“So if these novel neuromorphic
AI work as expected, would you believe one of these new machines to
possess phenomenal consciousness or 1PP?”



I think it is a possibility that some of this stuff could be running
directly on the intrinsic phenomenal qualities of the physics, and am
looking forward to hearing Colin’s thoughts on this.



Collin talks about the computation being done with “Natural EM fields”,
which are not included in the list of stuff used in this article, but that
could just be an oversight?



“All of the aforementioned large-scale neuromorphic computers are
silicon-based and implemented using conventional complementary metal oxide
semiconductor technology; however, there is a tremendous amount of research
in the neuromorphic community on developing new types of materials for
neuromorphic implementations, such as phase-change, ferroelectric,
non-filamentary, topological insulators or channel-doped biomembranes28
<https://www.nature.com/articles/s43588-021-00184-y#ref-CR28>,29
<https://www.nature.com/articles/s43588-021-00184-y#ref-CR29>,30
<https://www.nature.com/articles/s43588-021-00184-y#ref-CR30>. One popular
approach in the literature is using memristors as the fundamental device to
have resistive memory to collocate processing and memory31
<https://www.nature.com/articles/s43588-021-00184-y#ref-CR31>,32
<https://www.nature.com/articles/s43588-021-00184-y#ref-CR32>, but other
types of devices have also been used to implement neuromorphic computers,
including optoelectronic devices10
<https://www.nature.com/articles/s43588-021-00184-y#ref-CR10>. Each device
and material used to implement neuromorphic computers has unique operating
characteristics, such as how fast they operate, their energy consumption
and the level of resemblance to biology. The diversity of devices and
materials used to implement neuromorphic hardware today offers the
opportunity to customize the properties required for a given application.”



As they point out, all of this different physical stuff has different
intrinsic physical properties and behaviors.  The behavior of Intrinsic
phenomenal qualities of physics like redness, could be abstractly
described, as all the above different physics are being abstractly
described.  But without subjective dictionaries, you can’t know the
qualities you are describing.  In order to know the intrinsic qualities of
the physical stuff you are describing, you must directly apprehend it as it
is computationally bound, possibly done via any of these physical methods,
into your consciousness, so you, too can subjectively directly apprehend
the physical qualities being abstractly described, together with the rest
of your directly apprehended phenomenal conscious knowledge.


Note: Understanding that there are two ways to gain knowledge about the
intrinsic properties of physics, as described in this section of our video
<https://canonizer.com/videos/consciousness/?chapter=differentiate_reality_knowledge&t=302>
is a prerequisite to knowing what I'm attempting to say, above.





On Fri, Aug 12, 2022 at 8:54 PM Stuart LaForge via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

>
> Implementing neural networks as hardware on chips could do for
> training AI what ASICS did for bitcoin mining. Neuromorphic chips have
> a lot of potential IMO. Why simulate what you can instead reverse
> engineer?
>
> https://www.nature.com/articles/s43588-021-00184-y
>
> ----------------------------
> Abstract
>
> Neuromorphic computing technologies will be important for the future
> of computing, but much of the work in neuromorphic computing has
> focused on hardware development. Here, we review recent results in
> neuromorphic computing algorithms and applications. We highlight
> characteristics of neuromorphic computing technologies that make them
> attractive for the future of computing and we discuss opportunities
> for future development of algorithms and applications on these systems.
> ----------------------------
>
> In his recent Frontiers in Neuroscience article about phenomenal
> consciousness being mediated by the complex EM fields of the brain,
> Colin Hales wrote:
>
> "The creation of chip materials able to express EM fields structurally
> identical to those produced by neurons can be used to construct
> artificial neurons that replicate neuron signal processing through
> allowing the actual, natural EM fields to naturally interact in the
> manner they do in the brain, thereby replicating the same kind of
> signaling and signal processing (computation). This kind of in silico
> empirical approach is simply missing from the science." (Hales &
> Ericson, 2022)
>
> So Colin, it appears that the neuromorphic chips and computer
> architecture described in the Nature Computational Science article is
> exactly what you were suggesting right? So if these novel neuromorphic
> AI work as expected, would you believe one of these new machines to
> posses phenomenal consciousness or 1PP?
>
> Stuart LaForge
>
>
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20220813/a34f0dc8/attachment-0001.htm>