3 ms·

I just want to know when this thing is going to put me out of work forever with it’s insane and crazy math skills.

talldayo 2y ago
I don't think it has crazy math skills at all. That's what classical computing is really good at - give it a problem of arbitrary length and a computer should be able to string together instructions to yield you a sum. Quantum computing, especially right now, is simply focused on getting reliable input/output idempotency. It will be a really long time before it has insane and crazy math skills, and when it does, traditional CPU architectures will probably outperform it. TL:DR - if the Texas Instrument calculator didn't put you out of a job, neither will quantum computers.
- eastbound 2y ago
  Aren’t matrix calculations a perfect field for quantum computing? i.e. AI would progress extremely fast, wouldn’t it?
  - tmvphil 2y ago
    No, not really. There is no speed up for general matrix multiply, and generally it won't be practical for any problem with large amounts of input and output data. Closest thing is HHL algorithm for solving big linear systems, which requires a bunch of caveats on the matrix, and even then, it needs to be a subroutine of another quantum algorithm, since it outputs a quantum state, and not the full vector.
bawolff 2y ago
Unless your job involves factoring numbers into primes (or a few other limited math problems) probably never.
galaxyLogic 2y ago
There's two great lanes of progress happening in engineering these days: 1. Quantum Computing, and 2. AI. If they can find some synergies between the two, that could be THE major development. Make better AI by using Quantum Computers, and make better Quantum Computers by applying AI.
- chikere232 2y ago
  QC – nowhere near practical applications AI – being practically applied in many areas where it doesn't seem to be working well I really do hope these aren't our best lanes of progress
  - galaxyLogic 2y ago
    Here's an example of where AI does something of economic importance: https://www.seattletimes.com/business/technology/google-introduces-ai-agent-that-aces-15-day-weather-forecasts/ https://www.seattletimes.com/business/technology/google-intr... Accurate weather forecasts save money. Doing them on super-computers costs a lot of money. But with Google's AI solution that gets cheaper. As for QC I'm not saying it does much useful things yet, but companies are starting to use it and there is progress going on. What in your view are better lanes of progress than QC and AI?
    - chikere232 2y ago
      I sloppily used "AI" as a collective term for the currently hyped LLMs and image generators, where especially LLMs are being applied enthusiastically regardless if it works It's very true that there are many areas of ML where considerable progress has been made and is delivering value Are companies using QC? It's not like it can do much currently, as I understand it We're making a lot of progress all over the engineering side of computing. A lot of it is incremental, but a faster or more energy efficient processor is very practically useful in a way that QC isn't yet. We might well get optical interconnects on cpu dies before we get anything tangible from QC

I don't think it has crazy math skills at all. That's what classical computing is really good at - give it a problem of arbitrary length and a computer should be able to string together instructions to yield you a sum. Quantum computing, especially right now, is simply focused on getting reliable input/output idempotency. It will be a really long time before it has insane and crazy math skills, and when it does, traditional CPU architectures will probably outperform it. TL:DR - if the Texas Instrument calculator didn't put you out of a job, neither will quantum computers.

eastbound 2y ago
Aren’t matrix calculations a perfect field for quantum computing? i.e. AI would progress extremely fast, wouldn’t it?
- tmvphil 2y ago
  No, not really. There is no speed up for general matrix multiply, and generally it won't be practical for any problem with large amounts of input and output data. Closest thing is HHL algorithm for solving big linear systems, which requires a bunch of caveats on the matrix, and even then, it needs to be a subroutine of another quantum algorithm, since it outputs a quantum state, and not the full vector.

bawolff 2y ago

Unless your job involves factoring numbers into primes (or a few other limited math problems) probably never.

galaxyLogic 2y ago

There's two great lanes of progress happening in engineering these days: 1. Quantum Computing, and 2. AI. If they can find some synergies between the two, that could be THE major development. Make better AI by using Quantum Computers, and make better Quantum Computers by applying AI.

chikere232 2y ago
QC – nowhere near practical applications AI – being practically applied in many areas where it doesn't seem to be working well I really do hope these aren't our best lanes of progress
- galaxyLogic 2y ago
  Here's an example of where AI does something of economic importance: https://www.seattletimes.com/business/technology/google-introduces-ai-agent-that-aces-15-day-weather-forecasts/ https://www.seattletimes.com/business/technology/google-intr... Accurate weather forecasts save money. Doing them on super-computers costs a lot of money. But with Google's AI solution that gets cheaper. As for QC I'm not saying it does much useful things yet, but companies are starting to use it and there is progress going on. What in your view are better lanes of progress than QC and AI?
  - chikere232 2y ago
    I sloppily used "AI" as a collective term for the currently hyped LLMs and image generators, where especially LLMs are being applied enthusiastically regardless if it works It's very true that there are many areas of ML where considerable progress has been made and is delivering value Are companies using QC? It's not like it can do much currently, as I understand it We're making a lot of progress all over the engineering side of computing. A lot of it is incremental, but a faster or more energy efficient processor is very practically useful in a way that QC isn't yet. We might well get optical interconnects on cpu dies before we get anything tangible from QC