Is Analog Computing Making a Comeback in AI Research?

AI Thread Summary
The discussion centers on the potential resurgence of analog computing in AI research, particularly for self-learning AI. While analog computing utilizes real-time voltage and frequency changes for calculations, many argue that digital processors are more efficient for AI tasks due to their ability to handle non-linear functions more accurately. The initial step in creating an analog computer involves modeling circuits in SPICE before implementation. Although analog computing has historical advantages, its drawbacks have led to a preference for digital solutions in modern applications. Overall, the consensus leans towards digital methods being more practical for current AI needs.
The Art of DHT
Messages
1
Reaction score
0
TL;DR Summary
Anyone with experience building electronic analog computers
I am researching self-learning AI, and one of the most talked of solutions is returning to analog computing. Analog computing uses real time changes in voltage and frequency to run calculations or perform functions. Does anyone have any experience in building electronic analog computers based on the op amp?
 
Engineering news on Phys.org
Hello @The Art of DHT ,

:welcome: ##\qquad## !​

Yes, lots of folks have lots of experience :biggrin:.

That may answer your question, but my guess is that it doesn't help you very much :smile:.

google is your friend. Tried some terms too ?

What is it you want to do ?

##\ ##
 
The Art of DHT said:
I am researching self-learning AI, and one of the most talked of solutions is returning to analog computing.
That is the talk, but not the practice. It is faster to emulate the AI algorithm on a digital processor. Op-amps are good at linear operations, but AI requires non-linear functions, which are more quickly and more accurately emulated by the multiply-accumulate function in a digital processor.

The first step in building an analog computer is to model the analog circuit in SPICE. Only once the algorithm and circuit has been optimised should a parallel analog solution be implemented.
 
  • Like
Likes pbuk and BvU
The Art of DHT said:
AI ... one of the most talked of solutions is returning to analog computing.
The AI related analog computing (the one I know about) is not exactly the opamp kind, and honestly, I would rather take that as some kind of hybrid or unique solution than 'real' analog.

Analog computing had some strong points way back, but there was also serious drawbacks too.
By now it's simply more pain than gain.

I's just like the gold standard. Everybody knows why it was phased out, but still, some people just can't stop flirting with the idea.
 
  • Like
Likes Klystron and Hornbein
Thread 'Weird near-field phenomenon I get in my EM simulation'
I recently made a basic simulation of wire antennas and I am not sure if the near field in my simulation is modeled correctly. One of the things that worry me is the fact that sometimes I see in my simulation "movements" in the near field that seems to be faster than the speed of wave propagation I defined (the speed of light in the simulation). Specifically I see "nodes" of low amplitude in the E field that are quickly "emitted" from the antenna and then slow down as they approach the far...
Hello dear reader, a brief introduction: Some 4 years ago someone started developing health related issues, apparently due to exposure to RF & ELF related frequencies and/or fields (Magnetic). This is currently becoming known as EHS. (Electromagnetic hypersensitivity is a claimed sensitivity to electromagnetic fields, to which adverse symptoms are attributed.) She experiences a deep burning sensation throughout her entire body, leaving her in pain and exhausted after a pulse has occurred...
Back
Top