Saltar al contenido ForoHackers

A qubit is not "both 0 and 1": what the popular line gets wrong

Computación cuánticaInformación cuántica
#1144 · quantum_tunnel_rat · · Denunciar

The sentence "a qubit is both 0 and 1 at the same time" is everywhere, and it hides the interesting part. Here is a slightly more careful version.

A qubit's state is a pair of complex numbers, the amplitudes: a|0> + b|1>, with |a|² + |b|² = 1. When you measure, you get 0 with probability |a|² and 1 with probability |b|². You always get one classical bit out. Nothing is read out "both".

Where the power comes from, if it comes from anywhere: amplitudes can be negative or complex, so they can cancel. A quantum algorithm is a way to arrange the computation so that the amplitudes of wrong answers cancel and the amplitudes of right answers add up. That is interference, not parallel universes doing your homework.

Two more things that get blurred:

  • n qubits have 2^n amplitudes, but a measurement gives you n bits. You cannot read out the whole state, so "it stores exponentially much information" is misleading.
  • Most problems have no known quantum speed-up. The ones that do are special, and the speed-up is usually not "instant".

I am not a pessimist. I just think the field is interesting enough without the slogans.

#1896 · bitflip_betty · · Denunciar

Also worth saying: amplitudes are not probabilities that you can sample before the end. If you peek in the middle of the computation, you collapse the state and the interference is gone. That is half of why this hardware is so hard to build.

#1897 · overfit_otter · · Denunciar

Statistician hat on: "the amplitudes can cancel" is the line that made it click for me. Probabilities cannot be negative, so classical randomness can never do that. Is that the cleanest way to say what is different?

#1898 · quantum_tunnel_rat · · Denunciar

It is the cleanest short version I know. A random classical algorithm adds probabilities, which are non-negative, so contributions only accumulate. Amplitudes carry a sign or phase, so contributions can subtract. The rest is engineering to make that useful.

#1899 · lambda_llama · · Denunciar

Can I add that a measurement is a function from a state to a distribution over outcomes? I will see myself out.

Inicia sesión o únete para responder.