Quantum annealing debugging systems and methods
Computational systems and methods employ characteristics of a quantum processor determined or sampled between a start and an end of an annealing evolution per an annealing schedule. The annealing evolution can be reinitialized, reversed or continued after determination. The annealing evolution can be interrupted. The annealing evolution can be ramped immediately prior to or as part of determining the characteristics. The annealing evolution can be paused or not paused immediately prior to ramping. A second representation of a problem can be generated based at least in part on the determined characteristics from an annealing evolution performed on a first representation of the problem. The determined characteristics can be autonomously compared to an expected behavior, and alerts optionally provided and/or the annealing evolution optionally terminated based on the comparison. Iterations of annealing evolutions may be performed until an exit condition occurs.
1. A method of operation of a control system communicatively coupled to a quantum processor, the control system comprising at least one processor and at least one nontransitory processor-readable medium that stores at least one of processor-executable instructions or data, the method comprising:
determining, by the control system, an annealing evolution schedule to perform a quantum annealing evolution computation on a problem representation in a quantum processor;
causing, by the control system, a start of at least a first iteration of an annealing evolution of the quantum processor with the problem representation in accordance with the annealing evolution schedule; and
at least once during a performance of the first iteration of the annealing evolution:
determining, by the control system, at least one characteristic of the quantum processor at at least a first point in time between the start of the first iteration of the annealing evolution as specified by the annealing evolution schedule and before a completion of the first iteration of the annealing evolution as specified by the annealing evolution schedule; and
causing, by the control system, a change in a rate of the annealing evolution of the quantum processor from a rate specified by a previous annealing evolution schedule based at least on part on the determined at least one characteristic of the quantum processor.
2. The method of claim 1 wherein causing a change in a rate of annealing based at least on part on the determined at least one characteristic of the quantum processor includes decreasing a rate of the annealing evolution.
3. The method of claim 1 , further comprising:
determining that more time is needed to more thoroughly search for a ground state, and wherein causing a change in a rate of annealing based at least on part on the determined at least one characteristic of the quantum processor includes decreasing a rate of the annealing evolution in response to the determination that more time is needed to more thoroughly search for the ground state.
4. The method of claim 1 wherein causing a change in a rate of annealing based at least on part on the determined at least one characteristic of the quantum processor includes increasing a rate of the annealing evolution.
5. The method of claim 1 , further comprising:
determining that less time is needed in order to pass through an avoided crossing and intentionally evolve to an excited state, wherein causing a change in a rate of annealing based at least on part on the determined at least one characteristic of the quantum processor includes increasing a rate of the annealing evolution in response to the determination that less time is needed.
6. A system, comprising:
a quantum processor;
a control system communicatively coupled to the quantum processor, the control system comprising at least one processor and at least one nontransitory processor-readable medium that stores at least one of processor-executable instructions or data which, when executed by the at least one processor, causes the at least one processor to:
determine an annealing evolution schedule to perform a quantum annealing evolution computation on a problem representation in a quantum processor;
cause a start of at least a first iteration of an annealing evolution of the quantum processor with the problem representation in accordance with the annealing evolution schedule; and
at least once during a performance of the first iteration of the annealing evolution:
determine at least one characteristic of the quantum processor at at least a first point in time between the start of the first iteration of the annealing evolution as specified by the annealing evolution schedule and before a completion of the first iteration of the annealing evolution as specified by the annealing evolution schedule; and
cause a change in a rate of the annealing evolution of the quantum processor from a rate specified by a previous annealing evolution schedule based at least on part on the determined at least one characteristic of the quantum processor.
7. The method of claim 1 , wherein determining the at least one characteristic of the quantum processor comprises reading out respective states of one or more qubits.
8. The method of claim 7 , wherein reading out respective states of one or more qubits comprises reading out one or more detectors coupled respectively to the one or more qubits.
9. The method of claim 1 , wherein determining the at least one characteristic of the quantum processor at at least a first point in time between the start of the first iteration of the annealing evolution and before a completion of the annealing evolution schedule comprises determining the at least one characteristic of the quantum processor at a point s in the annealing evolution, and further comprises:
gathering a set of samples S(s) at the point s;
for each qubit q,
analyzing a set of statistics of the set of samples S(s) at the point s with respect to the qubit q;
determining a function F q (s) that represents an average value of the qubit q at the point s;
for each pair of qubits q, q′,
analyzing a set of statistics of the set of samples S(s) at the point s with respect to a product q·q′ of the pair of qubits q, q′;
determining a function G q q′ (s) that represents an average value of the product q·q′ at the point s; and
analyzing the function F q and the function G q q′ .
10. The method of claim 9 wherein:
for each function F q (s) that approaches ±1, causing, by the control system, a change in a rate of the annealing evolution from a rate specified by a previous annealing evolution schedule comprises delaying the rate in the annealing evolution for the qubit q; and
for each function G q q′ (s) that approaches 1, causing, by the control system, a change in a rate of the annealing evolution from a rate specified by a previous annealing evolution schedule comprises delaying the rate in the annealing evolution for the pair of qubits q,q′.
11. The system of claim 6 wherein the control system decreases a rate of the annealing evolution.
12. The system of claim 6 wherein the control system further determines that more time is needed to more thoroughly search for a ground state, and the control system decreases a rate of the annealing evolution in response to the determination that more time is needed to more thoroughly search for the ground state.
13. The system of claim 6 wherein the control system increases a rate of the annealing evolution.
14. The system of claim 6 wherein the control system further determines that less time is needed in order to pass through an avoided crossing and intentionally evolve to an excited state, and the control system increases a rate of the annealing evolution in response to the determination that less time is needed.
15. The system of claim 6 , wherein the at least one characteristic of the quantum processor comprises respective states of one or more qubits.
16. The system of claim 15 , wherein respective states of one or more qubits are detected by one or more detectors coupled respectively to the one or more qubits.
17. The system of claim 6 , wherein the at least a first point in time between the start of the first iteration of the annealing evolution and before a completion of the annealing evolution schedule comprises a point s in the annealing evolution, and wherein the at least one of processor-executable instructions or data cause the at least one processor to:
gather a set of samples S(s) at the point s;
for each qubit q,
analyze a set of statistics of the set of samples S(s) at the point s with respect to the qubit q;
determine a function F q (s) that represents an average value of the qubit q at the point s;
for each pair of qubits q, q′,
analyze a set of statistics of the set of samples S(s) at the point s with respect to a product q·q′ of the pair of qubits q, q′;
determine a function G q q′ (s) that represents an average value of the product q·q′ at the point s; and
analyze a function F q and the function G q q′ .
18. The system of claim 17 wherein:
for each function F q (s) that approaches 1, the change in the rate of the annealing evolution comprises a delay of the rate in the annealing evolution for the qubit q; and
for each function G q q′ (s) that approaches 1, the change in the rate of the annealing evolution comprises a delay of the rate in the annealing evolution for the pair of qubits q, q′.