Numerical quantum experimentation

1. A computer implemented method for numerical quantum experimentation, comprising:

receiving data identifying (i) a computational problem that is a candidate for a quantum computation, and (ii) one or more classical numerical algorithms for solving the computational problem;

providing input task data identifying (i) the computational problem, and (ii) the one or more classical numerical algorithms, to a numerical quantum experimentation system, wherein the numerical quantum experimentation system comprises multiple universal numerics workers, each universal numerics worker of the multiple universal numerics workers being configured to solve the computational problem using the one or more classical numerical algorithms;

receiving, from the numerical quantum experimentation system, data representing results of the one or more classical numerical algorithms to solve the computational problem, wherein the data comprises one or more measures relating to the classical numerical algorithms used to obtain solutions to the computational problem, the one or more measures comprising a total number of sweeps required by the classical numerical algorithms to obtain the solutions to the computational problem;

projecting the one or more measures relating to the classical numerical algorithms used to obtain solutions to the computational problem to estimated runtimes on respective quantum computation hardware; and

determining, using the received data and the estimated runtimes, whether the received data indicates that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem.

2. The method of claim 1 , further comprising, in response to determining that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem, performing a quantum computation to solve the computational problem.

3. The method of claim 1 , wherein determining that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem comprises determining that the computational problem experiences a quantum speed up when solved using quantum computation.

4. The method of claim 1 , wherein determining whether the received data indicates that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem comprises determining whether the received data indicates that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem based on one or more of:

computation runtime,

a scaling behavior of the computation runtime with respect to computational problem input size,

classical overheads of the quantum computation,

quantum hardware design complexity,

comparisons of runtime scaling for achieving approximate solutions at target levels of a target quality,

comparisons of obtaining a solution of at least a predetermined quality, wherein the predetermined quality represents an approximate solution to a problem.

5. The method of claim 1 , wherein determining whether the received data indicates that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem comprises:

determining, for each numerical algorithm used to solve the computational problem, an algorithm runtime; and

analyzing the algorithm runtimes to determine whether the received data indicates that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem.

6. The method of claim 5 , wherein the one or more classical numerical algorithms comprise (i) a classical simulation of a quantum system, and (ii) one or more classical algorithms, and

wherein analyzing the algorithm runtimes to determine whether the received data indicates that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem comprises comparing the runtime of the classical simulation of the quantum system to the runtimes of each of the one or more classical algorithms to determine whether the runtime of the classical simulation of the quantum system is shorter than the runtimes of each of the one or more classical algorithms.

7. The method of claim 1 , wherein determining whether the received data indicates that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem comprises:

analyzing the received data to extrapolate a scaling law for quantum computation runtime as a function of problem size; and

based on the scaling law, determining whether a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem.

8. The method of claim 1 , wherein determining whether the received data indicates that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem comprises:

using the received data to determine an approximability measure of quantum computation and classical computation; and

based on the determined approximability measure, determining whether the received data indicates that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem.

9. The method of claim 1 , wherein the universal numerics worker is configured to receive dynamical instructions from a numerical quantum experimentation system taskmaster to execute a particular algorithm of the one or more classical numerical algorithms through configuration received as part of an input task provided by a numerical quantum experimentation system client via the taskmaster.

10. The method of claim 1 , wherein the universal numerics worker comprises a numerics interface that shields the universal numerics worker from knowledge particular to any specific classical numerical algorithm.

11. The method of claim 1 , wherein the universal numerics worker comprises a respective cache of constructed numerics data objects from prior input tasks, and wherein the method further comprises, in response to receiving a new input task:

determining, by the universal numerics worker, whether the cache comprises data required for the new input task; and

in response to determining that the cache comprises data required for the new input task, retrieving and using the data required for the new input task.

12. The method of claim 1 , wherein

the computational problem comprises one or more of (i) classically simulating a quantum system or (ii) an abstract computational problem; and

the one or more classical numerical algorithms comprise one or more of (i) classical simulations of quantum systems, (ii) physically motivated classical algorithms, or (iii) abstract classical algorithms.

13. The method of claim 1 , wherein identifying one or more classical numerical algorithms for solving the candidate computational problem further comprises, for an identified classical numerical algorithm, identifying a respective set of parameter values for the algorithm.

14. The method of claim 1 , wherein receiving data representing results of the one or more classical numerical algorithms to solve the candidate computational problem comprises receiving data representing a result when the respective result is available.

15. The method of claim 1 , wherein data representing a result of a respective classical numerical algorithm to solve the computational problem comprises data representing respective computation time.

16. The method of claim 1 , wherein the universal numerics worker is configured to issue resource adjustment commands based on current load as measured by a number of tasks waiting for completion.

17. The method of claim 1 , wherein the numerical quantum experimentation system comprises multiple taskmasters, wherein a taskmaster of the multiple taskmasters is configured to manage universal numerics workers that are located within a predetermined distance to the taskmaster.

18. The method of claim 1 , wherein the numerical quantum experimentation system is configured to receive input tasks from multiple users of the numerical quantum experimental system.

19. A numerical quantum experimentation system, comprising:

a client;

one or more taskmasters;

multiple universal numerics workers;

wherein the numerical quantum experimentation system is configured to perform operations comprising:

receiving input task data identifying (i) a computational problem that is a candidate for a quantum computation, and (ii) one or more classical numerical algorithms for solving the computational problem, wherein a universal numerics worker, of the multiple universal numerics workers, is configured to solve the computational problem using the one or more classical numerical algorithms;

solving the computational problem using the one or more classical numerical algorithms;

providing, as output, data representing results of the one or more classical numerical algorithms to solve the computational problem, wherein the data comprises one or more measures relating to the classical numerical algorithms used to solve the computational problem, the one or more measures comprising a total number of sweeps required by the classical numerical algorithms to solve the computational problem;

projecting the one or more measures relating to the classical numerical algorithms used to obtain solutions to the computational problem to estimated runtimes on respective quantum computation hardware; and

determining, using the received data and the estimated runtimes, whether the received data indicates that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem.

20. The system of claim 19 , wherein the numerical quantum experimentation system further comprises a quantum computing device, and wherein the operations further comprise:

in response to determining that a quantum computation applied to the computational problem has a greater efficacy for arriving at a solution than a classical computation applied to the computational problem, performing a quantum computation using the quantum computing device to solve the computational problem.