You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Trying to compare performance of different experimental implementations of gates is a maze because different groups will often insist on reporting different quantities that are largely equivalent but not directly comparable.
Quantities such as entanglement fidelity, avg fidelity, RB decay rate, equivalent depolarization rate, \chi_00, etc, are all related by simple linear functions. For the sake of sanity, we should have functions that convert between all of these.
There are two possible approaches:
Function based: perhaps more natural, but not very composable, and will require many functions (for the 5 examples above, that's 20 functions). We'd need to define, e.g., avgfidelity2entfidelity, avgfidelity2rbdecay, etc.
Type based: will strike some as unfamiliar, but will require much less effort (we can choose a reference quantity, and then simply convert to and from this quantity, require only 5 constructors). e.g., AvgFidelity(f), RBDecayRate(r), EquivalentDepolRate(p). All types would inherent from AvgFidelityMetric, would maintain one of the equivalent metric internally (e.g., avg fidelity), along with the named metric.
The text was updated successfully, but these errors were encountered:
Trying to compare performance of different experimental implementations of gates is a maze because different groups will often insist on reporting different quantities that are largely equivalent but not directly comparable.
Quantities such as entanglement fidelity, avg fidelity, RB decay rate, equivalent depolarization rate, \chi_00, etc, are all related by simple linear functions. For the sake of sanity, we should have functions that convert between all of these.
There are two possible approaches:
avgfidelity2entfidelity
,avgfidelity2rbdecay
, etc.AvgFidelity(f)
,RBDecayRate(r)
,EquivalentDepolRate(p)
. All types would inherent fromAvgFidelityMetric
, would maintain one of the equivalent metric internally (e.g., avg fidelity), along with the named metric.The text was updated successfully, but these errors were encountered: