Meet Mythic AI’s Soon-to-be-Legendary Analog AI

I was rambling (as I sometimes do) about something or other a few weeks back, and I remarked that my buddy Jay Dowling had introduced me to a fascinating analogue computer called The Analog Thing (THAT). As I previously stated, it took me a long time to grasp that "THAT" was an abbreviation for "The Analog Thing," not some obscure element of the title.

Jay also directed me to a fantastic Veritasium YouTube video on analogue computers, which exhibits all kinds of amazing mechanical implementations of analogue functions, such as summing multiple sine waves. The mechanical analogue of integration left me speechless in admiration of our forefathers' genius.Derek Muller, the broadcaster, has one of the aforementioned THAT machines on his desk, which I only just noticed. He shows how to use this tiny beauty to replicate a damped mass vibrating on the end of a spring in the video (this is the sort of thing they had me doing as a university student way back in the mists of time).

But we're getting ahead of ourselves... Derek begins the film by describing how analogue computers were the most powerful computational engines on Earth for hundreds of years, forecasting eclipses and tides and commanding anti-aircraft weaponry.Derek goes on to say that the introduction of solid-state (semiconductor) electronics accelerated the adoption of digital computers, to the point that nearly every computer we use today is digital. "A perfect storm of variables is laying the stage for a revival of analogue technology," we're told next (I feel a drum roll would be appropriate at this point in the proceedings).

From beginning to end, this 22-minute film is well worth your time. If you're short on time, the description of how neurons in our brains function (on/off with weighting) that begins at 3:50 in the movie is a great way to visualise the foundation of the artificial neural networks (ANNs) we use to construct today's artificial intelligence.I'm afraid I've been a little sneaky, because once you've started watching this part, you'll be enthralled by the rest of the video, so you might as well do a Julie Andrews and "start at the very beginning — a very good place to start," and call it a day (the flash mob version of this in the central railway station in Antwerp, Belgium, always makes me smile with delight).

Did you know that the ResNet ANN has a "Top 5 Error Rate" of only 3.6 percent in 2015, which was better than the average human performance of 5.1 percent (see video for more details)?