-1

From various sources (one being this), I've heard that quantum computers find the correct answer in a sea of possible ones by destructively interfering the wrong ones with each other, and the right one keeps building on itself. How does this work? Does every wrong answers have an opposite, but still wrong, answer? How can the correct answer not? Or am I completely misled here?

Thanks

EDIT: This question differs form this because an answer to my previous question tells me about quantum interference to solve this problem, but does not explain how, and neither does the links.

Community
  • 1
APCoding
  • 323
  • 2
  • 9
  • 5
    What research have you done? We expect you to do a significant amount of research or self-study before asking, and to show us in the question what you've done. See also http://cs.stackexchange.com/help/how-to-ask. Within the past 24 hours you asked a related question and were recommended some resources where you can learn more. Have you read them? Those resources already have the answers to your questions. See also [tag:quantum-computing], e.g., http://cs.stackexchange.com/q/48045/755. – D.W. Oct 12 '15 at 23:42
  • @D.W. I have Googled this many times, as well as look at all the resources suggested in my previous answer – APCoding Oct 13 '15 at 00:51
  • 1
    There are no shortcuts. You will have to take the time and actually read about quantum computing. – Yuval Filmus Oct 13 '15 at 03:50

1 Answers1

2

It depends very strongly on the problem being solved. Some graphs are very amenable to quantum random walks; others aren't. Some operations work well with quantum phase estimation, but that technique isn't so useful for understanding what a black box is doing.

Grover's algorithm works based off of the fact that you can negate every amplitude in a system across the average of all the amplitudes. The amplitude of solutions is pushed away from the global average, and towards high values, by negating the phase of solutions (e.g. by conditioning on a predicate) before each average-flip.

Craig Gidney
  • 5,842
  • 22
  • 47