Feature details
It would be great to have some possibility to work natively with qudit wires, where the Hilbert space of each wire is of some dimension d . This is natural to some cold atom and ion experiments and nicely interpolates between qubit and CV systems.
Implementation
For qudits we have two native gate sets, some of which are already included in the pennylane-ls package:
- angular momentum operators, where on each wire the native operations would be
RLX, RLY, RLZ and RLZ2. They are basically rotations of gaussian states on the collective Bloch spheres and squeezing operation. They are common to cold atoms.
- generalized Pauli operators, which can be used for stabilizer experiments. The are native to ions.
- A translation between the two gate sets would be a next step, but seems highly non-trivial.
- It remains fairly unclear where such qudit operations and devices should be implemented.
How important would you say this feature is?
1: Not important. Would be nice to have.
Additional information
This would allow for gate based description of a number of experiments.
Ions:
https://arxiv.org/abs/2109.06903
Cold atoms:
https://arxiv.org/abs/1507.03782
https://arxiv.org/abs/2010.15923
Feature details
It would be great to have some possibility to work natively with qudit wires, where the Hilbert space of each wire is of some dimension d . This is natural to some cold atom and ion experiments and nicely interpolates between qubit and CV systems.
Implementation
For qudits we have two native gate sets, some of which are already included in the
pennylane-lspackage:RLX,RLY,RLZandRLZ2. They are basically rotations of gaussian states on the collective Bloch spheres and squeezing operation. They are common to cold atoms.How important would you say this feature is?
1: Not important. Would be nice to have.
Additional information
This would allow for gate based description of a number of experiments.
Ions:
https://arxiv.org/abs/2109.06903
Cold atoms:
https://arxiv.org/abs/1507.03782
https://arxiv.org/abs/2010.15923