Quantum Chess: Decomposition into CZ gates

[dstrain115]

Quantum chess curreently decomposes moves into circuits containing square root of iSWAP gates (ie. ISWAP ** 0.5). It would also be good to be able to decompose these operations into CZ-based gates to support future processors.

In particular, the methods in the following files need to be able to support outputting circuits using CZ:

https://github.com/quantumlib/ReCirq/blob/master/recirq/quantum_chess/quantum_moves.py
https://github.com/quantumlib/ReCirq/blob/master/recirq/quantum_chess/controlled_iswap.py

This would be a good project for a 20% contributor interested in learning about quantum computing.

[dstrain115]

This was some code that we used to run it on a CZ grid for one time. We might just want something more user-friendly.

import cirq
import cirq_google as cg
import unitary.quantum_chess.ascii_board as ab
import unitary.quantum_chess.move as m
import unitary.quantum_chess.quantum_board as qb
import unitary.quantum_chess.circuit_transformer as ct

sampler = cirq.Simulator()

gs = cirq.CZTargetGateset()
mapper=ct.ConnectivityHeuristicCircuitTransformer(device)
transformer = lambda circuit: mapper(cirq.optimize_for_target_gateset(circuit, gateset=gs))
board=qb.CirqBoard(
init_basis_state=0,
sampler=sampler,
device=device,
transformer=transformer
)
example_move=m.Move.from_string("g1^f3h3:SPLIT_JUMP:BASIC")
b = ab.AsciiBoard(board=board)
b.reset()
b.apply(example_move)
print(b)