Doug Strain
Doug Strain
The quantum chess concept docs still refer to large code blocks that are now in recirq. These should mostly (or completely) moved to the relevant pointers in recirq. For instance,...
(More difficult issues) One of the biggest limitations of quantum chess right now is the number of qubits and length of the circuit. The current quantum chess algorithm assumes that...
We should explore using calibration metrics (https://quantumai.google/cirq/google/calibration) as a way of picking better qubits for quantum chess (i.e. picking qubits with lower sq/sqrt-iswap/readout errors). See also quantumlib/unitary#51 and quantumlib/ReCirq#127 for...
During one of the demos (king face-off in game 1 of [quantum chess quarter finals](https://tachyon-618ltf726387nm8o2xv8v5kwt436sljm.twitch.tv/videos/824369168)), a bad qubit caused an embarrassing situation. It would be good if we could have...
En passant tests cannot be run with real devices: https://github.com/quantumlib/ReCirq/blob/master/recirq/quantum_chess/quantum_board_test.py#L777 Note that these tests are not marked with the pytest parameterize for ALL_CIRQ_BOARDS since they cannot be mapped properly to...
(More difficult issue) When we try to map to hardware (especially for merge moves), sometimes the moves do not fit the topology of the device. For instance, consider this move...
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...
- Part of #905. - Removed _t_complexity_ calculation and moved to assert since this has a call_graph, the T count can be inferred.
Fixes: #263
- Add descriptions and items for the perimeter zone for quantum rpg example.