In classical optimization problems, some objective function's surface serves as a landscape for finding its minimum through minimization algorithms.

Instead of your typical U-shaped objective surface (e.g. the squared loss function), consider instead a W-shaped surface, implying two pockets where one pocket must contain the global minimum, while the other only holds a local minimum.

Quantum annealing, in contrast to classical simulated annealing, has an additional ability to evaluate both pockets of the W-shape concurrently, from what I understand based on 'parallel universes' or quantum parallel computation that classical, sequential optimization can't. How much of this ability is due to superposition, entanglement and teleportation?

(It seems to me like superposition and entanglement both contribute to overall parallelizing behavior, but with one being an option reserved for certain applications, outside of, or such as, quantum annealing.)

Could someone also go into detail about the maths/circuitry behind this process, of how quantum annealing is designed to jump between or parallelize the optimizer in two or more different locations of the objective surface at once?

A referral to a Python package that demonstrates quantum annealing would be good too.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.