Important note - lasers are relevant in ion traps or neutral atom quantum computers. The biggest players are using superconducting as a base - which doesn't need lasers, but instead near absolute zero temperatures. Obviously that's insanely expensive, but also easier in some ways; so alternatives still have some hope.
However, replacing a simple PID loop or PDH with AI? Not sure if the extra complexity is needed; more so effort should be spent on making the technology cheaper and more accessible.
>However, replacing a simple PID loop or PDH with AI? Not sure if the extra >complexity is needed
Laser guy here. It's a highly-non-linear search problem (so not great for PID until you are close), and human intuition is used in surprising ways. For instance, maybe the operator knows that one adjuster is particularly 'wonky' and drifts more than the others; this is discovered through trial and error. Believe it or not, some people seem to have "the touch", and others never seem to get it (I think it might just be a matter of patience - alignment can be boring as hell and some people find it frustrating.) To be trained, usually someone starts out by "shadowing" a more experienced person, but eventually people develop their own style that can be hard to fully verbalize.
I am curious if they could de-tune the system, ask the AI to hit a target, and get anything sensible. Because the AI would be relentless, I bet it might work better than expected...
I've worked with neutral atom QCs before (though am not a laser tech): keeping lasers locked is a massive pain in the ass.
> more so effort should be spent on making the technology cheaper and more accessible
Lasers are weird because the techniques to generate light at different frequencies differs a lot. "Telecom wavelengths" used in, say fiber optic networking, is super cheap and widely available, but these QCs often need different wavelengths that aren't mass produced. Some systems even need the ability to tune the wavelength precisely and the range of that tuning is another spec that can matter. I'm sure these QC companies would love for this to be more stable and cheaper, but the reality is without demand (of which there is little), it's just not going to happen.
Important note - lasers are relevant in ion traps or neutral atom quantum computers. The biggest players are using superconducting as a base - which doesn't need lasers, but instead near absolute zero temperatures. Obviously that's insanely expensive, but also easier in some ways; so alternatives still have some hope.
However, replacing a simple PID loop or PDH with AI? Not sure if the extra complexity is needed; more so effort should be spent on making the technology cheaper and more accessible.
>However, replacing a simple PID loop or PDH with AI? Not sure if the extra >complexity is needed
Laser guy here. It's a highly-non-linear search problem (so not great for PID until you are close), and human intuition is used in surprising ways. For instance, maybe the operator knows that one adjuster is particularly 'wonky' and drifts more than the others; this is discovered through trial and error. Believe it or not, some people seem to have "the touch", and others never seem to get it (I think it might just be a matter of patience - alignment can be boring as hell and some people find it frustrating.) To be trained, usually someone starts out by "shadowing" a more experienced person, but eventually people develop their own style that can be hard to fully verbalize.
I am curious if they could de-tune the system, ask the AI to hit a target, and get anything sensible. Because the AI would be relentless, I bet it might work better than expected...
I've worked with neutral atom QCs before (though am not a laser tech): keeping lasers locked is a massive pain in the ass.
> more so effort should be spent on making the technology cheaper and more accessible
Lasers are weird because the techniques to generate light at different frequencies differs a lot. "Telecom wavelengths" used in, say fiber optic networking, is super cheap and widely available, but these QCs often need different wavelengths that aren't mass produced. Some systems even need the ability to tune the wavelength precisely and the range of that tuning is another spec that can matter. I'm sure these QC companies would love for this to be more stable and cheaper, but the reality is without demand (of which there is little), it's just not going to happen.
If I read the article correctly,
- This involve multiple variable, so it is never a simple PID loop
- They didn't use LLM for the feedback loop, they use the LLM to test and fine tune their program.