Maybe I'm just incompetent today. But what PCM devices did they use? Internal IBM research devices? Or did they only use the simulator (https://github.com/IBM/aihwkit)? Are there commercial PCM chips yet?
On 10/1/21 7:34 AM, Marcus Daniels wrote: > Companies like Mythic and IBM have developed analog devices for energy- > efficient deep learning. Noise and low precision are often used as part of > ML training protocols anyway. Here they did careful side-by-side testing > to quantify the impact of going analog. > > https://www.frontiersin.org/articles/10.3389/fncom.2021.675741/full > <https://www.frontiersin.org/articles/10.3389/fncom.2021.675741/full> > > Or perhaps the proposition is that CS people (for example) can’t appreciate > complexity that where the generative rules are unknown or obscure? Sure > that horse is fast, but how can I put a new engine in it? -- "Better to be slapped with the truth than kissed with a lie." ☤>$ uǝlƃ .-- .- -. - / .- -.-. - .. --- -. ..--.. / -.-. --- -. .--- ..- --. .- - . FRIAM Applied Complexity Group listserv Zoom Fridays 9:30a-12p Mtn UTC-6 bit.ly/virtualfriam un/subscribe http://redfish.com/mailman/listinfo/friam_redfish.com FRIAM-COMIC http://friam-comic.blogspot.com/ archives: 5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/ 1/2003 thru 6/2021 http://friam.383.s1.nabble.com/