With $n$ qubits, $2^n$ features can be encoded with amplitude embedding, while only $n$ with angle embedding. So, is there any reason to use angle embedding?
Asked
Active
Viewed 344 times
1
-
what's your definition/understanding of amplitude and angle encoding? – glS Mar 07 '23 at 17:39
-
@glS see https://docs.pennylane.ai/en/stable/code/api/pennylane.AmplitudeEmbedding.html and https://docs.pennylane.ai/en/stable/code/api/pennylane.AngleEmbedding.html, for example – salcc Mar 07 '23 at 21:04
1 Answers
2
One of the main advantage of angle embedding is that it can be performed in a constant time with parralelism : each qubit will go through a rotation gate in parralel.
Using only one gate to encode each qubit reduces the noises which are important in today's NISQ computers.

Duen
- 326
- 1
- 10