| Popis: |
In this paper, we prove the equivalence between sampling advice, i.e., advice in the form of a training set, and classical advice. Specifically, our main result demonstrates that BPP/samp is equal to P/poly. Additionally, we delve into the analysis of these relationships under the constraint of a fixed distribution. Notably, we show that under such circumstances, the equality does not hold. This result remains valid when considering quantum advice and a quantum generalization of the training set. Finally, leveraging the insights gained from these proofs, we identify sufficient and necessary complexity assumptions for the existence of concept classes that exhibit a quantum learning speed-up in the worst-case scenario, i.e., when accurate results are required for all inputs. |
