Validated that near-term quantum hardware can deliver measurable advantages, rather than benefits being decades away.
pull down to refresh
related posts
36 sats \ 2 replies \ @SimpleStacker 9h
Smells extremely fishy to me. To be quite frank, I don't believe it at all. My guess: an unscrupulous or overly enthusiastic research lead has pulled the wool over an oblivious executive's eyes.
Curious about @south_korea_ln's take.
In fact, I will put my money where my mouth is and short some quantum stocks.
reply
0 sats \ 1 reply \ @siggy47 OP 9h
When I saw this I knew I had to post it to get some solid push back. Putting quantum aside, the predictive improvement seems exaggerated- "up to 34% improvement."
reply
36 sats \ 0 replies \ @SimpleStacker 8h
I know quite a bit about predictive models, and a little bit about quantum (I was also a physics major in college.)
I can't really understand where a quantum computer would help in the predictive model. The only part I can think of is that it can speed up the process of estimating the model parameters. But this would normally be used for a speed up of finding the best parameters... it shouldn't change the performance once the best parameters have been found.
So unless the parameters that the classical computer was finding is incredibly worse than the parameters that the quantum computer is finding, or unless they are arbitrarily limiting the amount of time you're allowed to use for finding the parameters... then it doesn't make much sense why performance would increase by that much.
Also, a 34% improvement on a model that "predicts the probability that an order gets filled at a given price" is hard to interpret. How crappy was your original model that it can be improved by 34%? For example, if the original accuracy was 90%, what does a 34% performance increase even mean?
A lot of this just smells off.
reply
36 sats \ 0 replies \ @optimism 8h
I say prove it with a reproducible result, or I call BS. lol
reply
36 sats \ 0 replies \ @kepford 9h
Been talking about this at work. Many are skeptical. Most actually. Great way to pump a stock though. I need evidence
reply