Comments by "Luis Aldamiz" (@LuisAldamiz) on "Memristors for Analog AI Chips" video.

  1. I'm intrigued but I fail to fully understand the concept and the implications.
    4
  2. Didn't Huawei just announce they made one like a few weeks or rather months ago?
    3
  3.  @RedRouge-j4j  - Chaos always wins in the end: small errors > big consequences.
    2
  4.  @ゾカリクゾ  - TY. Is it the same idea as that of physical (hardware) neural networks that I've seen somehwere that is a (potential, under research) alternative to virtual (software) ones, which are the ones being used now in AIs?
    1
  5.  @timng9104  TY for the reply. However I recall that one characteristic of these brain-like (under research) physical neural networks was also to imitate the way brains work, which is "slow", "low energy" and "chaotic" (as opposed to the heavily centralized CPU architectures). This is, I understand, not necessarily what a "simple" shift from transistor (CPU & RAM) to memristor architecture but retaining the current mainline "neural network" paradigm would do, right? IDK, probably there are several parallel and somewhat related lines of research here, all cutting-edge and thus a bit hard to understand. If they succeed, they may be behind a new AI and overall computing revolution (for the good and for the bad). Or it may not work at all, like the alleged graphene revolution and such.
    1
  6.  @RedRouge-j4j  - Nah, I don't remember: I was pretty much off the loop of computer tech developments between the early 90s and the last few years. Somehow I managed to still own a computer all this time... but I was focusing on other stuff (from prehistory and genetics to fundamental physics) and didn't bother checking all the developmens, which I'm now updating to but with many lacunae. Anyway, the issue of "niche applications" may be like those of GPUs, which have become much more than just graphics (and very relevant re. AI). IMHO if these kind of developments succeed technologically, they have radical altering properties for all the computer industry. And they don't look like that long tale of quantum computing, which alwas spooked me as one of those physical practical impossibilities along with fussion energy: this seems more conceptually sound, even if surely needing some real development, consolidation on something that actually does something useful (and revolutionary).
    1