Quantum streaming kernel

1. A method comprising:

obtaining input data for a computer process configured to receive an input;

by operation of one or more classical processors, parameterizing quantum logic gates based on the input data;

executing a quantum logic circuit on a quantum resource, wherein the quantum resource defines input qubits, intermediate qubits, and output qubits, and executing the quantum logic circuit comprises:

applying the parameterized quantum logic gates to the input qubits;

applying multi-qubit quantum logic gates that couple the input qubits with one or more of the intermediate qubits;

applying multi-qubit quantum logic gates that couple one or more of the intermediate qubits with the output qubits; and

measuring the output qubits to obtain an output of the quantum logic circuit;

generating output data based on the output of the quantum logic circuit; and

providing the output data as the input for the computer process.

2. The method of claim 1 , comprising executing multiple iterations of an iterative process, wherein each iteration comprises:

obtaining an input data set for the iteration;

parameterizing the quantum logic gates for the iteration based on the input data set for the iteration;

executing, on the quantum resource, the quantum logic circuit comprising the quantum logic gates parameterized for the iteration; and

generating an output data set for the iteration based on the execution of the quantum logic gates parameterized for the iteration.

3. The method of claim 2 , comprising:

receiving an input data stream from a remote source;

obtaining the input data set for each iteration from the input data stream;

generating an output data stream comprising the output data set for each iteration; and

providing the output data stream as the input for the computer process.

4. The method of claim 2 , wherein the quantum resource maintains a coherent state of the input qubits and the intermediate qubits between consecutive iterations of the iterative process.

5. The method of claim 2 , wherein one or more of the iterations comprises:

obtaining an input data set for the iteration;

obtaining an output data set from a prior iteration; and

parameterizing the quantum logic gates for the iteration based on the input data set for the iteration and the output data set from the prior iteration.

6. The method of claim 2 , wherein each iteration comprises:

applying, to the input qubits, the quantum logic gates parameterized for the iteration;

applying the multi-qubit quantum logic gates that couple the input qubits with one or more of the intermediate qubits;

applying the multi-qubit quantum logic gates that couple one or more of the intermediate qubits with the output qubits; and

applying a set of measurement operations to the output qubits.

7. The method of claim 1 , wherein the parameterized quantum logic gates comprise single-qubit rotation gates, and parameterizing the quantum logic gates comprises defining rotation angles of the respective single-qubit rotation gates.

8. The method of claim 1 , wherein the quantum logic gates comprise controlled-phase gates, and parameterizing the quantum logic gates comprises defining phases of the respective controlled-phase gates.

9. The method of claim 1 , wherein:

applying the multi-qubit quantum logic gates that couple the input qubits with one or more of the intermediate qubits comprises applying a first set of two-qubit entangling gates, each of the first set of two-qubit entangling gates being applied to a respective pair of qubits comprising one of the input qubits and one of the intermediate qubits; and

applying the multi-qubit quantum logic gates that couple one or more of the intermediate qubits with the output qubits comprises applying a second set of two-qubit entangling gates, each of the second set of two-qubit entangling gates being applied to a respective pair of qubits comprising one of the intermediate qubits and one of the output qubits.

10. The method of claim 1 , wherein measuring the output qubits to obtain an output of the quantum logic circuit comprises obtaining a bitstring representing a computational state of the output qubits, and the output data comprises the bitstring.

11. The method of claim 1 , wherein measuring the output qubits to obtain an output of the quantum logic circuit comprises obtaining a raw measurement comprising analog data.

12. The method of claim 1 , wherein executing the quantum logic circuit comprises:

applying multi-qubit quantum logic gates that couple one or more of the intermediate qubits with other intermediate qubits; and

applying single-qubit quantum logic gates to the input qubits, the intermediate qubits, and the output qubits.

13. A computer system comprising:

one or more classical processors;

one or more quantum processors that define input qubits, intermediate qubits, and output qubits; and

memory storing instructions that, when executed by the one or more classical processors, cause the one or more classical processors to perform operations comprising:

parameterizing quantum logic gates based on input data;

obtaining an output produced by the one or more quantum processors executing a quantum logic circuit, wherein executing the quantum logic circuit comprises:

applying the parameterized quantum logic gates to the input qubits;

applying multi-qubit quantum logic gates that couple the input qubits with one or more of the intermediate qubits;

applying multi-qubit quantum logic gates that couple one or more of the intermediate qubits with the output qubits; and

measuring the output qubits to obtain the output; and

generating output data based on the output of the quantum logic circuit.

14. The computer system of claim 13 , wherein the operations comprising:

obtaining the input data for a computer process configured to receive an input; and

providing the output data as the input for the computer process.

15. The computer system of claim 13 , wherein the one or more quantum processors comprises an array of devices that includes:

an input layer comprising qubit devices that define the input qubits;

an intermediate layer comprising qubit devices that define the intermediate qubits; and

an output layer comprising qubit devices that define the output qubits.

16. The computer system of claim 15 , wherein the array comprises a two-dimensional array, the input layer comprises a first row of the qubit devices, the output layer comprises a second row of the qubit devices and the intermediate layer comprises one or more rows of the qubit devices between the first row and the second row.

17. The computer system of claim 15 , wherein the array comprises a three-dimensional array.

18. The computer system of claim 13 , wherein the one or more quantum processors comprises an array of connected devices that can be controlled simultaneously and that defines the input qubits, the intermediate qubits and the output qubits, and wherein the output qubits can be measured without measuring the input qubits or the intermediate qubits.

19. A method comprising converting an input stream of data to an output stream of data by repeatedly:

receiving a new portion of the input stream of data;

after encoding the new portion into an internal quantum state of a quantum processor, measuring a first part of the internal quantum state of the quantum processor while maintaining coherence of a second part of the internal quantum state of the quantum processor; and

producing a new portion of the output stream of data based on the measurement of the first part of the internal quantum state.

20. The method of claim 19 , wherein maintaining the coherence of the second part of the internal quantum state preserves information from a history of the input stream of data, and each measurement contains information based on the history of the input stream of data.

21. The method of claim 19 , wherein the quantum processor comprises interconnected qubits that define the internal quantum state, and wherein measuring a first part of the internal quantum state of the quantum processor while maintaining coherence of a second part of the internal quantum state of the quantum processor comprises measuring a first subset of the interconnected qubits.

22. The method of claim 19 , wherein the input stream of data has a structured time dependence.

23. The method of claim 22 , wherein the input stream of data comprises a time series of data sets, and each new portion of the input stream of data comprises a new data set from the time series.

24. The method of claim 23 , wherein a coherence time of the internal quantum state is longer than a time between data sets in the time series.

25. The method of claim 19 , wherein converting the input stream to the output stream comprises preprocessing the input stream, and the method comprises providing the output stream of data as an input for a computing process.

26. The method of claim 19 , wherein converting the input stream to the output stream comprises encoding each new portion into the internal quantum state of the quantum processor by:

parameterizing quantum logic gates based on the new portion of the input stream of data; and

executing a quantum logic circuit on the quantum processor, wherein the quantum processor defines input qubits, intermediate qubits, and output qubits, and executing the quantum logic circuit comprises:

applying the parameterized quantum logic gates to the input qubits;

applying multi-qubit quantum logic gates that couple the input qubits with one or more of the intermediate qubits; and

applying multi-qubit quantum logic gates that couple one or more of the intermediate qubits with the output qubits.

27. The method of claim 26 , wherein measuring the first part of the internal quantum state of the quantum processor while maintaining the second part of the internal quantum state of the quantum processor comprises, after executing the quantum logic circuit, measuring the output qubits without measuring the input qubits or the intermediate qubits.