Systems and methods for enabling access to physics-inspired computers and physics-inspired computer simulators

Abstract

A computing system and method are disclosed for enabling a processing device to remotely access a computing platform including at least one physics-inspired computer simulator having variable parameters. The computing system includes a communication interface for receiving a request, the request including at least one computational task for processing using the physics-inspired computer simulator having variable parameters, a control unit operable to convert the request into instructions for the at least one physics-inspired computer simulator, transmit the instructions, and receive at least one corresponding solution, and a memory for storing one or more of the at least one computational task, a data set, the variable parameters, and the at least one corresponding solution.

Description

[Technical Field]

[0001] One or more embodiments of the present invention are directed to a computing system and method for providing remote access to a computing platform over a network, which in one or more embodiments allows a user to choose between using an actual physics-inspired computer and using a simulator thereof. In particular, one or more embodiments of the computing system enable the use of a physics-inspired computer simulator instead of an actual physics-inspired computer, reducing costs. [Background technology]

[0002] Today, the scientific community has devised many different noisy intermediate-scale quantum (NISQ) devices, as well as other physics-inspired devices and computers, that are constantly being developed, improved, and published. Although optimization tasks, stochastic sampling, and / or other computational tasks can be performed significantly faster due to the variety of quantum and / or other physics phenomena, access to these machines remains prohibitively expensive for categories of users such as research groups and startups. This situation is caused by several factors, including high manufacturing and maintenance costs.

[0003] A need is recognized herein for a method and / or system that overcomes at least one of the limitations associated with access to such computers. Summary of the Invention

[0004] According to a broad aspect, a computing system is disclosed for enabling a processing device to remotely access a computing platform over a network, the computing platform including at least one physics-inspired computer simulator including variable parameters. The computing system includes: a communication interface for receiving a request provided by a processing unit, the request including at least one computational task for processing using at least one physics-inspired computer simulator including variable parameters; a control unit operatively connected to the communication interface and the at least one physics-inspired computer simulator including variable parameters, the control unit for converting the received request into instructions for the at least one physics-inspired computer simulator, transmitting the instructions to the at least one physics-inspired computer simulator to perform the at least one computational task, and receiving at least one corresponding solution; and a memory operatively connected to the control unit and the at least one physics-inspired computer simulator, the memory for storing one or more of the at least one computational task, a data set included in the received request, variable parameters of the at least one physics-inspired computer simulator, and the received at least one corresponding solution.

[0005] According to one or more embodiments, the computing platform further includes at least one physics-inspired computer. The request further includes a selection of at least one of the at least one physics-inspired computer simulator and the at least one physics-inspired computer. Furthermore, the control unit is further operably connected to the at least one physics-inspired computer. If the selection of the request is a selection of the at least one physics-inspired computer, the control unit is further used for converting the received request into instructions for the at least one physics-inspired computer and transmitting the instructions to the at least one physics-inspired computer to perform the at least one computational task. Furthermore, the control unit is further used for receiving at least one corresponding solution from the at least one physics-inspired computer.

[0006] According to one or more embodiments, the computing system further includes a training unit operatively connected to the at least one physics-inspired computer simulator, the training unit for training the at least one physics-inspired computer simulator.

[0007] According to one or more embodiments, the computing system includes a training unit operatively connected to at least one physics-inspired computer simulator, the training unit for training the at least one physics-inspired computer simulator.

[0008] According to one or more embodiments, the at least one physics-inspired computer simulator type corresponds to a physics-inspired computer, and the training unit is used to train the at least one physics-inspired computer simulator using at least the instructions transmitted to the physics-inspired computer and using at least one corresponding solution obtained from the physics-inspired computer.

[0009] According to one or more embodiments, the communication interface receives the plurality of requests. Additionally, the computing platform further includes a queuing unit for queuing the received plurality of requests according to a criterion.

[0010] According to one or more embodiments, if the corresponding request includes an indication that the at least one computational task and at least one corresponding solution are usable for training purposes, the training unit trains the at least one physics-inspired computer simulator using the corresponding instructions transmitted to the physics-inspired computer and the at least one corresponding solution obtained from the physics-inspired computer.

[0011] According to one or more embodiments, a processing device for remotely accessing a computing system includes a digital computer operably connected to a communication interface via a data network.

[0012] According to one or more embodiments, the instructions for the physics-inspired computer and the instructions for the at least one physics-inspired computer simulator are identical.

[0013] According to one or more embodiments, the computing platform includes a distributed computing system.

[0014] According to one or more embodiments, a physics-inspired computer includes a non-classical computer.

[0015] According to one or more embodiments, the non-classical computer is selected from the group consisting of a NISQ device, a quantum computer, a superconducting quantum computer, an ion trap quantum computer, quantum annealing, an optical quantum computer, a spin-based quantum dot computer, and a photonic-based quantum computer.

[0016] According to one or more embodiments, at least one physics-inspired computer simulator includes a computer-implemented method that includes: mimicking a physics-inspired computer output for a given input; using a training unit to update at least one of the variable parameters, thereby improving corresponding performance; Includes:

[0017] According to one or more embodiments, the training unit is selected from the group consisting of a tensor processing unit (TPU), a graphical processing unit (GPU), a field programmable gate array (FPGA), and an application specific integrated circuit (ASIC).

[0018] According to one or more embodiments, at least one physics-inspired computer simulator includes a neural network.

[0019] According to a broad aspect, a computer-implemented method for enabling remote access to a computing platform including at least one physics-inspired computer simulator having variable parameters is disclosed. The method includes receiving, at a communication interface, a request, the request including at least one computational task for processing using the at least one physics-inspired computer simulator, converting the at least one computational task of the received request into instructions suitable for the at least one physics-inspired computer simulator, providing the instructions to the at least one physics-inspired computer simulator, receiving at least one corresponding generated solution resulting from execution of the instructions, and providing the at least one corresponding generated solution.

[0020] According to one or more embodiments, the computing system further includes at least one physics-inspired computer and a training unit for training the at least one physics-inspired computer simulator. Furthermore, the received request further includes an indication of a selection of at least one of the physics-inspired computer and the at least one physics-inspired computer simulator for use in processing the at least one computational task. Furthermore, the at least one computational task of the request is converted into instructions suitable for at least one of the physics-inspired computer and the at least one physics-inspired computer simulator. Furthermore, providing the instructions is performed in at least one of the physics-inspired computer and the at least one physics-inspired computer simulator in response to the indication of the selection.

[0021] According to one or more embodiments, the request is received from a digital computer operably connected to the communication interface using a data network, and at least one corresponding generated solution is provided to the digital computer.

[0022] According to one or more embodiments, the at least one corresponding generated solution is obtained from at least one physics-inspired computer simulator, and further includes training at least one physics-inspired computer simulator using the at least one corresponding generated solution and the at least one computational task.

[0023] According to one or more embodiments, training is performed if the request includes an indication that at least one computational task and at least one corresponding generated solution are available for training purposes.

[0024] According to one or more embodiments, training includes executing a procedure based on a machine learning protocol using at least one corresponding generated solution and at least one computational task, and updating variable parameters of a physics-inspired computer simulator accordingly.

[0025] According to one or more embodiments, the method further includes storing the instructions and at least one corresponding generated solution.

[0026] An advantage of one or more embodiments of the methods and computing systems disclosed herein is that they enable access to physics-inspired computer simulators trained using real computational tasks, such as simulators of quantum devices, which are relatively cheaper than access to quantum devices.

[0027] Another advantage of one or more embodiments of the methods and computing systems disclosed herein is that they allow for the emulation of physics-inspired computers such as quantum computers.

[0028] Another advantage of one or more embodiments of the methods and computing systems disclosed herein is that they enable the use of at least one computational task to improve a physics-inspired computer simulator.

[0029] In order that one or more embodiments of the invention may be readily understood, one or more embodiments of the invention are shown by way of example in the accompanying drawings. [Brief explanation of the drawings]

[0030]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

[0031] In the following description of one or more embodiments, references to the accompanying drawings are made to illustrate examples in which the invention may be practiced.

[0032] <Terminology> The term "invention" and the like means "one or more inventions disclosed in this application" unless otherwise specified.

[0033] The terms "aspect," "one embodiment," "embodiment," "this embodiment," "this embodiment," "one or more embodiments," "an embodiment," "particular embodiment," "one embodiment," "another embodiment," and the like mean "one or more, but not all, embodiments of the disclosed invention," unless expressly stated otherwise.

[0034] Reference to "another embodiment" or "another aspect" when describing an embodiment does not imply that the referenced embodiment is mutually exclusive with another embodiment (e.g., an embodiment described before the referenced embodiment), unless specifically stated otherwise.

[0035] The terms "include," "comprise," and variations thereof mean "including but not limited to," unless otherwise specified.

[0036] The terms "indefinite article," "definite article," and "at least one" mean "one or more," unless otherwise specified.

[0037] The term "plurality" means "two or more" unless otherwise specified.

[0038] The term "herein" means "in this application, including anything that may be incorporated by reference," unless otherwise specified.

[0039] The term "whereby" is used herein only before a clause or other group of words that express only the intended result, purpose, or result of something expressly recited above. Thus, when the term "whereby" is used in a claim, the clause or other word that it modifies does not establish any specific further limitations on the claim or limit the meaning or scope of the claim.

[0040] The term "for example" and similar terms mean "for example," and thus do not limit the term or phrase they describe. For example, in the sentence "A computer transmits data (e.g., instructions, data structures) over the Internet," the term "for example" explains that "instructions" are an example of "data" that a computer may transmit over the Internet, and also explains that "data structures" are an example of "data" that a computer may transmit over the Internet. However, both "instructions" and "data structures" are merely examples of "data," and things other than "instructions" and "data structures" can be "data."

[0041] The term "ie" and similar terms mean "in other words" and thus qualify the term or phrase they describe.

[0042] When values ​​are described as ranges, it will be understood by those of skill in the art that such disclosure includes disclosure of all possible subranges within such ranges, as well as specific numerical values ​​falling within such ranges, whether or not a specific numerical value or specific subrange is explicitly stated.

[0043] In the following detailed description, reference is made to the accompanying drawings, which form a part of this specification. In the drawings, like symbols typically identify like components unless the context dictates otherwise. The exemplary embodiments described in the detailed description, drawings, and claims are not meant to be limiting. Other embodiments may be utilized, and other changes may be made, without departing from the scope of the subject matter presented herein. It will be readily understood that aspects of the present disclosure, as generally described herein and illustrated in the drawings, can be arranged, substituted, combined, separated, and designed in a wide variety of different configurations, all of which are expressly contemplated herein.

[0044] As used herein, the term "classical," when used in the context of calculation or computation, generally refers to calculations performed using binary values ​​that use discrete bits, without the use of quantum mechanical superposition and quantum mechanical entanglement. A classical computer can be a digital computer, such as a computer that uses discrete bits (e.g., 0 and 1), without the use of quantum mechanical superposition and quantum mechanical entanglement.

[0045] As used herein, the term "non-classical," when used in the context of computing or computation, generally refers to any method or system for performing computational procedures outside the paradigm of classical computing.

[0046] As used herein, the term "physics-inspired," when used in the context of calculation or computation, generally refers to any method or system for performing computational procedures that are based at least in part on and / or mimic any physical phenomenon.

[0047] As used herein, the term "quantum device" generally refers to any device or system that performs computations using any quantum mechanical phenomenon, such as quantum mechanical superposition and quantum mechanical entanglement.

[0048] As used herein, the terms "quantum computing," "quantum procedure," "quantum operation," and "quantum computer" generally refer to any method or system for performing computations using quantum mechanical operations (such as unitary transformations on quantum channels or completely positive trace-preserving (CPTP) maps) on a Hilbert space represented by a quantum device.

[0049] As used herein, the term "quantum computer simulator" generally refers to any computer-implemented method using any classical hardware that provides solutions to computational tasks, mimicking the results provided by a quantum computer.

[0050] As used herein, the term "physics-inspired computer simulator" generally refers to any computer-implemented method using any classical hardware that provides solutions to computational tasks, mimicking the results provided by a physics-inspired computer.

[0051] As used herein, the term "noisy intermediate-scale quantum device" (NISQ) generally refers to any quantum device that can perform tasks that exceed the capabilities of today's classical digital computers.

[0052] The present disclosure discloses one or more embodiments of a method and computing system for enabling access to at least one of at least one physics-inspired computer and at least one physics-inspired computer simulator in a distributed computing environment.

[0053] Neither the title nor the abstract should be construed as limiting in any way, as well as the scope of the disclosed invention. The title of this application and the section headings provided herein are for convenience only and should not be construed as limiting the disclosure in any way.

[0054] Numerous embodiments are described in this application and are presented for illustrative purposes only. The described embodiments are not, and are not intended to be, limiting in any sense. The invention of the present disclosure is broadly applicable to numerous embodiments, as will be readily apparent from the present disclosure. Those skilled in the art will recognize that the invention of the present disclosure can be implemented with various modifications and alterations, such as structural and logical modifications. While certain features of the invention of the present disclosure may be described with reference to one or more specific embodiments and / or drawings, it will be understood that such features are not limited to use in the one or more specific embodiments or drawings in which they are described, unless otherwise specified.

[0055] It will be understood that one or more embodiments of the present invention can be implemented in many ways. These implementations, or any other form the present invention may take, may be referred to herein as systems or techniques. Components such as processors or memories described as being configured to perform a task include either general-purpose components temporarily configured to perform the task at a given time, or specific components manufactured to perform the task.

[0056] With all of this in mind, one or more embodiments of the present invention are directed to a method and computing system for providing remote access over a network to a computing platform, the computing platform including at least one physics-inspired computer simulator that includes variable parameters.

[0057] One or more embodiments of the present invention enable access to at least one of at least one physics-inspired computer and at least one physics-inspired computer simulator based on machine learning algorithms.

[0058] It will be appreciated that the physics-inspired computer simulator aims to provide a cost-effective alternative to quantum computers or other physics-inspired computers, allowing users to accelerate computations compared to their classical counterparts at a relatively low price. In one or more embodiments, a user has the option to choose between a physics-inspired computer and a physics-inspired computer simulator based on their requirements, and submits the problem to the selected solver.

[0059] It will be appreciated that in one or more embodiments, the posed problems are decoded into instructions suitable for a physics-inspired computer, as described further below. In one or more embodiments, these instructions are directed to either a physics-inspired computer or a physics-inspired computer simulator. Furthermore, if the instructions are directed to an actual physics-inspired computer, the posed problems, along with the results, can be further used to improve the performance of the physics-inspired computer simulator using machine learning techniques.

[0060] For example, a quantum annealing simulator can include a conditional generative model. The conditional generative model is pre-trained using samples obtained from either an existing sampling algorithm, such as Metropolis-Hastings Monte Carlo, or an actual quantum device, such as quantum annealing. The conditional generative model can generate corresponding data samples. The data samples are further used to solve the original problem posed by the user.

[0061] <Physics-inspired computers> It will be understood that a physics-inspired computer may include one or more of optical computing devices such as optical parametric oscillators (OPOs) and integrated photonic coherent Ising machines, quantum computers such as quantum annealing, or implementations of physics-inspired methods such as gate-based quantum computers, simulated annealing, simulated quantum annealing, population annealing, and quantum Monte Carlo.

[0062] <Quantum device> Any type of quantum computer may be suitable for the techniques disclosed herein. In accordance with the description herein, suitable quantum computers include, by way of non-limiting example: Superconducting quantum computers (qubits implemented as tiny superconducting circuits (Josephson junctions)) (Clarke et al., "Superconducting quantum bits," Nature , Issue 453 , No. 7198 , pp. 1031-1042, (2008) ) Ion trap quantum computer (qubits implemented as states in ion traps) (Kielpinski et al., "Architecture for a large-scale ion-trap quantum computer," Nature , Issue 417 , No. 6890 , pp. 709-711, (2002) ) Optical lattice quantum computers (qubits implemented as the states of neutral atoms trapped in an optical lattice) (Deutsch et al., "Quantum computing with neutral atoms in an optical lattice," arXiv preprint: quant-ph / 0003022 (2000)) Spin-based quantum dot computers (qubits implemented as trapped electron spin states) (Imamoglu et al., "Quantum information processing using quantum dot spins and cavity QED," arXiv preprint: quant-ph / 9904096 (1999)) Spatial-based quantum dot computers (qubits implemented as electron positions in double quantum dots) (Fedichkin et al., "Novel coherent quantum bit using spatial quantization levels in semiconductor quantum dots," arXiv preprint: quant-ph / 0006097 (2000)) Coupled quantum wires (qubits implemented as pairs of quantum wires coupled by quantum point contacts) (Bertoni et al., "Quantum logic gates based on coherent electron transport in quantum wires," Physical Review Letters , Issue 84, No. twenty five , pg. 5912, (2000) ) Nuclear magnetic resonance quantum computers (qubits implemented as nuclear spins and probed with radio waves) (Cory et al., "Nuclear magnetic resonance spectroscopy: An experimentally accessible paradigm for quantum computing," arXiv preprint: quant-ph / 9709001 (1997)) Solid-state NMR Kane quantum computer (qubits implemented as nuclear spin states of phosphorus donors in silicon) (Kane, BE, "A silicon-based nuclear spin quantum computer," Nature, Issue 393, No .6681 , pp. 133-137, (1998) ) Electrons-on-helium quantum computer (qubits implemented as electron spin) (Lyon, SA, "Spin-based quantum computing using electrons on liquid helium," arXiv preprint: cond-mat / 0301581 (2006)) Quantum computers based on cavity quantum electrodynamics (qubits implemented as trapped atomic states coupled to highly precise resonators) (Burell, Z., "An Introduction to Quantum Computing using Cavity QED concepts," arXiv preprint: arXiv:1210.6512 (2012)) Quantum computers based on molecular magnets (qubits implemented as spin states) (Leuenberger et al., "Quantum computing in molecular magnets," arXiv preprint: cond-mat / 0011415 (2001)) Fullerene-based ESR quantum computers (qubits implemented as electron spins of atoms or molecules wrapped in fullerenes) (Harneit, W., " Spin Quantum Computing with Endohedral Fullerenes," arXiv preprint:arXiv: 1708.09298 (2017)) Linear optical quantum computers (qubits implemented as processing states of different modes of light via linear optical elements such as mirrors, beam splitters, and phase shifters) (Knill et al., "Efficient linear optics quantum computation," arXiv preprint: quant-ph / 0006088 (2000)) Diamond-based quantum computers (qubits implemented as electron or nuclear spins in NV centers in diamond) (Nizovtsev et al., "A quantum computer based on NV centers in diamond: optically detected nutations of single electron and nuclear spins," Optics and S spectroscopic , Issue 99, No. 2 , pp. 233-244 , (2005) ) Quantum computers based on Bose-Einstein condensates (qubits implemented as two component BECs) (Byrnes et al., "Macroscopic quantum computation using Bose-Einstein condensates," arXiv preprint: quantum-ph / 1103.5512 (2011)) Transistor-based quantum computers (qubits implemented as semiconductors coupled to photonic resonators) (Sun et al., "A single-photon switch and transistor enabled by a solid-state quantum memory," arXiv preprint: quant-ph / 1805.01964 (2018)) Quantum computers based on rare-earth ion-doped inorganic crystals (qubits implemented as hyperfine levels of the atomic ground state of rare-earth ion-doped inorganic crystals) (Ohlsson et al., "Quantum computer hardware based on rare-earth-ion-doped inorganic crystals," Optics C communications , Issue 201, No. 1-3 , pp. 71-77 , (2002) ) Quantum computers based on metal-like carbon nanospheres (qubits implemented as electron spins in conductive carbon nanoparticles) (Nafradi et al., "Room temperature manipulation of long lifetime spins in metallic-like carbon nanospheres," arXiv preprint: cond-mat / 1611.07690 (2016)) D-Wave Quantum Annealing (Quantum Bits Implemented as Superconducting Logic Elements) (Johnson et al., "Quantum annealing with manufactured spins," Nature , Issue 473, No. 7346 , pp. 194-198 , (2011) )

[0063] <Noisy Intermediate-Scale Quantum Technology> The term "Noisy Intermediate-Scale Quantum Technology (NISQ)" was introduced by John Preskill ("Quantum Computing in the NISQ era and beyond," arXiv preprint: arXiv:1801.00862). Here, "Noisy" means that the control over qubits is imperfect, and "Intermediate-Scale" refers to the number of qubits, which can range from 50 to several hundred. Several physical systems made from superconducting qubits, artificial atoms, and ion traps have been proposed as viable candidates for building both NISQ quantum devices and ultimately universal quantum computers.

[0064] <Quantum Annealing> Quantum annealing is a quantum mechanical system consisting of multiple manufactured qubits.

[0065] A bias source called a local field bias is inductively coupled to each qubit. In one embodiment, the bias source is an electromagnetic device used to pass magnetic flux through the qubit in order to provide control over the state of the qubit (see U.S. Patent Application No. 2006 / 0225165).

[0066] The local field bias on the qubit is programmable and controllable. In one or more embodiments, a qubit control system including a digital processing unit is connected to the qubit system to enable programming and adjustment of the local field bias on the qubit.

[0067] Quantum annealing can further include multiple couplings between multiple pairs of multiple qubits. In one embodiment, the coupling between two qubits is a device near both qubits that passes magnetic flux to both qubits. In the same embodiment, the coupling can consist of a superconducting circuit interrupted by a compound Josephson junction. Magnetic flux can pass through the compound Josephson junction, thereby passing magnetic flux to both qubits (see U.S. Patent Application Publication No. 2006 / 0225165). The strength of this magnetic flux contributes quadratically to the energy of a quantum Ising model with a transverse magnetic field. In one embodiment, the coupling strength is forced by adjusting a coupling device near both qubits.

[0068] The coupling strengths can be controllable and programmable. In one or more embodiments, a quantum annealing control system including a digital processing unit is connected to the multiple couplings and is capable of programming the quantum annealing coupling strengths.

[0069] In one or more embodiments, quantum annealing performs a transformation of the quantum Ising model with a transverse magnetic field from an initial configuration to a final configuration, in such embodiments, the initial and final configurations of the quantum Ising model with a transverse magnetic field provide a quantum system described by their corresponding initial and final Hamiltonians.

[0070] It will be appreciated that quantum annealing can be used as a heuristic optimizer of these energy functions. An embodiment of such an analog processor is disclosed by McGeoch et al. (“Experimental Evaluation of an Adiabatic Quantum System for Combinatorial Optimization,” Computing Frontiers; May 14-16, 2013) and also in U.S. Patent Application No. 2006 / 0225165.

[0071] It will be appreciated that quantum annealing can further be used to provide samples from the Boltzmann distribution of the corresponding Ising model at finite temperatures (Bian et al. , " The Ising model: teaching an old problem new tricks , " (2010) , and Amin et al. , " Quantum Boltzmann Machine ," arXiv preprint: arXiv:1601.02036 (2016) ). This method of sampling is called quantum sampling.

[0072] <Optical Computing Device> Another embodiment of an analog system that can sample the Boltzmann distribution of the Ising model near its equilibrium state is an optical device.

[0073] In one or more embodiments, the optical device includes a network of optical parametric oscillators (OPOs) such as those disclosed in U.S. Patent Application Publication No. 2016 / 0162798 and International Application Publication No. 2015006494.

[0074] In this embodiment, each spin in the Ising model is simulated by an optical parametric oscillator (OPO) operating at degeneracy.

[0075] A degenerate optical parametric oscillator (OPO) is an open dissipative system that experiences a second-order phase transition at the oscillation threshold. Due to phase-sensitive amplification, a degenerate OPO can oscillate with either a zero or a π phase relative to the pump phase for amplitudes above the threshold. The phase is random and is affected by quantum noise associated with optical parametric down-conversion during oscillation buildup. Therefore, a degenerate OPO naturally represents a binary number specified by its output phase. Based on this property, a degenerate OPO system can be used as a physical representative of an Ising spin system. The phase of each degenerate OPO is identified as an Ising spin, and its amplitude and phase are determined by the strength and sign of the Ising coupling between the associated spins.

[0076] When pumped by a strong source, a degenerate optical parametric oscillator (OPO) adopts one of two phase states corresponding to spin +1 or -1 in the Ising model. A network of N substantially identical optical parametric oscillators (OPOs) with mutual coupling is pumped with the same source to simulate an Ising spin system. After a transient period from the introduction of the pump, the network of OPOs approaches a steady state near its thermal equilibrium.

[0077] The phase state selection process relies on vacuum fluctuations and the cross-coupling of optical parametric oscillators (OPOs). In some embodiments, the pump is modulated at a constant amplitude. In other embodiments, the pump power is increased incrementally. In still other embodiments, the pump is controlled in other ways.

[0078] In one or more embodiments of the optical device, multiple couplings of the Ising model are simulated by multiple configurable couplings used to couple optical fields between optical parametric oscillators (OPOs). The configurable couplings can be configured to be off and configured to be on. Turning the couplings on and off can be done gradually or abruptly. When configured to be on, the configurations can provide any phase or amplitude depending on the coupling strength of the Ising model.

[0079] Each OPO output is interfered with using a phase reference and the resulting signal is captured at a photodetector. The OPO outputs represent the configuration of the Ising model. For example, in the Ising model, zero phase represents the spin -1 state and π phase represents the +1 spin state.

[0080] For the Ising model with spin, according to one or more embodiments, the resonant cavities of multiple optical parametric oscillators (OPOs) are configured to have a round-trip time equal to twice the period of the pulses from the pump source. Round-trip time, as used herein, refers to the time it takes light to propagate along one of the described recursive paths. Pulses of a pulse train having a period equal to the period of the round-trip time of the resonant cavities can propagate through the optical parametric oscillators (OPOs) simultaneously without interfering with each other.

[0081] In one or more embodiments, coupling of the optical parametric oscillator (OPO) is provided by multiple delay lines distributed along the resonant cavity.

[0082] The multiple delay lines contain multiple modulators that synchronously control the strength and phase of the coupling, allowing the optical device to be programmed to simulate the Ising model.

[0083] In a network of optical parametric oscillators (OPOs), a delay line and a corresponding modulator are sufficient to control the amplitude and phase of the coupling between every two OPOs.

[0084] In one or more embodiments, an optimizer capable of sampling from an Ising model can be fabricated as a network of optical parametric oscillators (OPOs), such as those disclosed in U.S. Patent Application Publication No. 2016 / 0162798.

[0085] In one or more embodiments, the network of optical parametric oscillators (OPOs) and the coupling of the optical parametric oscillators (OPOs) are achieved using commercially available mode-locked lasers and optical elements, such as telecommunications fiber delay lines, modulators, and other optical devices. Alternatively, the network of optical parametric oscillators (OPOs) and the coupling of the optical parametric oscillators (OPOs) are implemented using optical fiber technology, such as fiber technology developed for telecommunications applications. It will be appreciated that the coupling can be achieved using fiber and controlled by an optical Kerr shutter.

[0086] <Integrated Photonic Coherent Ising Machine> Another embodiment of an analog system capable of sampling from the Boltzmann distribution of an Ising model near its equilibrium state is the unified photonic coherent Ising machine disclosed, for example, in U.S. Patent Application Publication No. 2018 / 0267937.

[0087] In one or more embodiments, an integrated photonic coherent Ising machine is a combination of nodes and a connection network that solves a particular Ising problem. In such embodiments, the combination of nodes and a connection network can form an optical computer that is adiabatic. In other words, the combination of nodes and a connection network solves the Ising problem non-deterministically when the values ​​stored in the nodes reach a steady state and minimize the energy of the nodes and the connection network. The values ​​stored in the nodes at the minimum energy level can be associated with values ​​that solve the particular Ising problem. The probabilistic solution can be sampled from a Boltzmann distribution defined by a Hamiltonian corresponding to the Ising problem.

[0088] In such an embodiment, the system includes a plurality of ring resonator photonic nodes. Each one of the plurality of ring resonator photonic nodes stores a value. A pump is coupled to each one of the plurality of ring resonator photonic nodes via a pump waveguide to provide energy to each one of the plurality of ring resonator photonic nodes. A connection network includes a plurality of pairs of elements. Each pair of elements includes a plurality of phase shifters for tuning the connection network with parameters associated with an encoding of the Ising problem. The connection network processes the value stored in each one of the plurality of ring resonator photonic nodes. The Ising problem is solved at a minimum energy level by the value stored in each one of the plurality of ring resonator photonic nodes.

[0089] <Digital Annealing> It will be understood that in one or more embodiments, digital annealing refers to a digital annealing unit, such as those developed by Fujitsu®.

[0090] <Algorithms implemented using quantum devices> It will be understood that any type of algorithm that can be implemented in a quantum device may be suitable for one or more embodiments of the methods and computing systems disclosed herein. In accordance with the description herein, suitable algorithms include, by way of non-limiting example, the following: Variational Quantum Eigensolver (VQE), a scalable co-design framework for solving chemical problems on quantum computers (Peruzzo et al. , " A variational eigenvalue solver on a q uantum processor,” 2013, arXiv preprint: arXiv: 1304.3061, and Nam et al. , " Ground-state energy estimation of the water molecule on a trapped ion quantum computer, " a rXiv preprint: arXiv:1902.10171 (2019) ) Grover's algorithm (Chuang et al.) is a quantum algorithm that allows quadratic speedup compared to its classical counterpart for search tasks. , " Experimental implementation of fast quantum searching, " P hysical R review L etters ; 1998, 80(15), p.340 8) The Deutsch-Josas algorithm, an efficient quantum algorithm for solving the Deutsch-Josas problem (Jones et al. , " Implementation of a quantum algorithm on a nuclear magnetic resonance quantum computer, " T the Journal of C chemicalP hysics , 1998 , 109(5), pp.1648-1653, arXiv preprint: arXiv:quant-ph / 9801027, and Debnath et al. , " Demonstration of a small programmable quantum computer with atomic qubits, "N ature, 536(7614), p.63 , arXiv preprint: arXiv:1603.04512 (2016) ) Shor's algorithm (Lu et al.) is a quantum algorithm for integer factorization that can exponentially speed up classical state-of-the-art factorization algorithms. , " Demonstration of a compiled version of Shor's quantum factoring algorithm using photonic qubits, " P Physical Review Letters ; 2007 , 99(25), p.250504 , arXiv preprint: arXiv:0705.1684, and Monz et al. , " Realization of a scalable Shor algorithm, " S science, 2016, 351(6277), pp.1068-1070 , arXiv preprint: arXiv:1507.08852 (2015) )

[0091] <Physics-inspired computer simulator> It will be understood that any type of simulator and simulation of a physics-inspired computer may be suitable for one or more embodiments of the methods and computing systems disclosed herein. It will be understood that a physics-inspired simulator may be any computer-implemented method using any classical hardware that provides a solution to a computational task, mimicking the results provided by a physics-inspired computer. A physics-inspired simulator is based on an artificial intelligence method. A physics-inspired simulator includes, for example, any machine learning method. Any machine learning method may be, for example, a supervised machine learning method and an unsupervised machine learning method, both of which may be combined with a reinforcement learning method. A physics-inspired simulator may include any reinforcement learning method.

[0092] In one or more embodiments, the quantum computer simulator is represented by a probabilistic framework, in which reinforcement learning of parameters defining a generative machine learning model, specifically a Restricted Boltzmann Machine, is performed to obtain a neural network representation of the ground state and time-dependent physical states of a given quantum Hamiltonian. The network weights should generally be complex-valued, providing a complete description of both the amplitude and phase of the wave function. The neural network parameters are optimized (trained in the language of neural networks) by static variational Monte Carlo sampling, or by time-dependent variational Monte Carlo if dynamical properties are of interest. For more details, see Carleo et al. , " Solving the quantum many-body problem with artificial neural networks ," Science , 2017 , 355(6325), pp.602-606, arXiv preprint: arXiv:1606.02318 (2016) , and Melko et al. , "Restricted Boltzmann machines in quantum physics, " N Art Physics , 2019 , 15(9), pp.887-892, and Torlai et al. , " Neural-network quantum state tomography, " N Art Physics , Issue 14, No. 447 , arXiv preprint: arXiv:1703.05334 (2017) Please refer to.

[0093] In one or more other embodiments, the quantum computer simulator is represented by a Deep Boltzmann Machine. This configuration proves capable of representing the exact ground states of a large class of many-body lattice Hamiltonians. In such an embodiment, two layers of hidden neurons propagate quantum correlations between the physics degrees of freedom in the visible layer. This approach reproduces the exact imaginary-time Hamiltonian expansion and is fully deterministic. A compact and accurate network representation of the ground states is obtained without stochastic optimization of network parameters. Physical quantities can be measured by sampling configurations of both the physics degrees of freedom and the neuronal degrees of freedom. For more details, see Carleo et al. , " Constructing exact representations of quantum many-body systems with deep neural networks, " N ature C Communications, 9(1), p.5322 , 2018 Refer to.

[0094] In one or more alternative embodiments, the quantum computer simulator includes a recurrent neural network (more precisely, a structure consisting of several stacked gated recurrent units, or GRUs). Using this type of scalable machine learning procedure allows for the reconstruction of both pure and mixed states. It will be appreciated that this method is experimentally friendly, since it only requires measurements of the quantum system. The learning procedure comes with a built-in approximate proof of reconstruction and makes no assumptions about the purity of the state under investigation. It can efficiently handle a wide variety of complex systems, including archetypal states in quantum information systems, as well as ground states of local spin models common in condensed matter physics. The procedure involves reducing state tomography to an unsupervised learning problem of the statistics of quantum measurements. This constitutes a state-of-the-art machine learning approach to the validation of complex quantum devices. It can further prove relevant as a neural network quantum circuit (Ansatz) over mixed states suitable for variational optimization. For a detailed description, see Carrasquilla et al. , " Reconstructing quantum states with generative models, " N ature Machine Intelligence, 1 (3), p.155, arXiv preprint: arXiv:1810.10584 , 2019 Refer to.

[0095] It will be appreciated that the quantum computer simulator may include a generative model that is trained on the quantum state tomography data. The generative model may include a neural network that represents the quantum states.

[0096] A neural network can be used as a functional representation of a wave function that describes a quantum state. Those skilled in the art will appreciate that neural network quantum state tomography is one possible process for training a neural network quantum state.

[0097] Those skilled in the art will appreciate that quantum state tomography (QST), the use of measurements to reconstruct quantum states, is the gold standard for validating and benchmarking quantum devices (Cramer et al., "Efficient quantum state tomography," Nature Communications , Issue 1 , No. 1, (2010)). The number of measurements and time required to accurately reconstruct the state using QST scale exponentially with the size of the system. In neural network tomography, JPEG0007741802000001.jpg951 is reconstructed from a set of measurements of the system. This strategy maps the learned probability distribution of a neural network to a probabilistic representation of the wave function.

[0098] <Digital Computer> In one or more embodiments, the digital computer includes one or more hardware central processing units (CPUs) that perform the functions of the digital computer. In one or more embodiments, the digital computer further includes an operating system configured to execute the executable instructions. In one or more embodiments, the digital computer is connected to a computer network. In one or more embodiments, the digital computer is connected to the Internet to access the World Wide Web. In one or more embodiments, the digital computer is connected to a cloud computing infrastructure. In one or more embodiments, the digital computer is connected to an intranet. In one or more embodiments, the digital computer is connected to a data storage device.

[0099] Those skilled in the art will appreciate that various types of digital computers may be used. Indeed, suitable digital computers include, by way of non-limiting example, server computers, desktop computers, laptop computers, notebook computers, subnotebook computers, netbook computers, set-top computers, media streaming devices, handheld computers, Internet appliances, mobile smartphones, tablet computers, personal digital assistants, video game devices, and vehicles. Smartphones may be suitable for use with one or more embodiments of the methods and systems described herein. Select televisions, video players, and digital music players, possibly with computer network connectivity, may be suitable for use with one or more embodiments of the systems and methods described herein. Suitable tablet computers may include booklets, slates, and tablet computers with convertible configurations.

[0100] In one or more embodiments, a digital computer includes an operating system. The operating system is configured to execute executable instructions. The operating system can be software, including, for example, programs and data, that manages the device's hardware and provides services for the execution of applications. Those skilled in the art will appreciate that various types of operating systems can be used. Indeed, suitable server operating systems include, by way of non-limiting example, FreeBSD, OpenBSD, NetBSD®, Linux, Apple®, Mac OS X Server®, Oracle®, Solaris®, Windows Server®, and Novell®, NetWare®. Suitable personal computer operating systems can include, by way of non-limiting example, UNIX-like operating systems such as Microsoft®, Windows®, Apple®, Mac OS X®, UNIX®, and GNU / Linux®. In one or more embodiments, the operating system is provided by cloud computing. Suitable mobile smartphone operating systems may include, by way of non-limiting example, Nokia®, Symbian® OS, Apple® iOS®, Research In Motion®, BlackBerry OS®, Google® Android®, Microsoft® Windows Phone® OS, Microsoft® Windows Mobile® OS, Linux®, and Palm® WebOS®. Suitable media streaming device operating systems may include, by way of non-limiting example, Apple TV®, Roku®, Boxee®, Google TV®, Google Chromecast®, Amazon Fire®, and Samsung® HomeSync®.Suitable video game device operating systems may include, by way of non-limiting example, Sony® PS3®, Sony® PS4®, Microsoft® Xbox 360®, Microsoft Xbox One, Nintendo® Wii®, Nintendo® Wii U®, and Ouya®.

[0101] In one or more embodiments, the digital computer includes a storage and / or memory device. Those skilled in the art will appreciate that various types of storage and / or memory may be used in a digital computer. In one or more embodiments, the storage and / or memory device includes one or more physical devices used to temporarily or permanently store data or programs. In one or more embodiments, the device includes volatile memory, requiring power to maintain stored information. In one or more embodiments, the device includes non-volatile memory, retaining stored information even when power is not supplied to the digital computer. In one or more embodiments, the non-volatile memory includes flash memory. In one or more embodiments, the non-volatile memory includes dynamic random access memory (DRAM). In one or more embodiments, the non-volatile memory includes ferroelectric random access memory (FRAM). In one or more embodiments, the non-volatile memory includes phase change random access memory (PRAM). In one or more embodiments, the device includes a storage device. Storage devices include, by way of non-limiting example, CD-ROMs, DVDs, flash memory devices, magnetic disk drives, magnetic tape drives, optical disk drives, and cloud computing-based storage devices. In one or more embodiments, the storage and / or memory device includes a combination of devices such as those disclosed herein.

[0102] In one or more embodiments, the digital computer includes a display device. A display device is used to provide visual information to a user. Those skilled in the art will appreciate that various types of display devices can be used. In one or more embodiments, the display device includes a cathode ray tube (CRT). In one or more embodiments, the display device includes a liquid crystal display (LCD). In one or more embodiments, the display device includes a thin film transistor liquid crystal display (TFT-LCD). In one or more embodiments, the display device includes an organic light emitting diode (OLED) display. In one or more embodiments, the OLED display device includes a passive matrix OLED (PMOLED) or an active matrix OLED (AMOLED) display. In one or more embodiments, the display device includes a plasma display. In one or more embodiments, the display device includes a video projector. In one or more embodiments, the display device includes a combination of devices, such as the devices disclosed herein.

[0103] In one or more embodiments, the digital computer includes an input device to receive information from a user. Those skilled in the art will appreciate that various types of input devices may be used. In one or more embodiments, the input device includes a keyboard. In one or more embodiments, the input device includes a pointing device. Pointing devices include, by way of non-limiting example, a mouse, a trackball, a trackpad, a joystick, a game controller, or a stylus. In one or more embodiments, the input device includes a touchscreen or multi-touchscreen. In one or more embodiments, the input device includes a microphone for capturing voice or other sound input. In one or more embodiments, the input device includes a video camera or other sensor for capturing motion or visual input. In one or more embodiments, the input device includes a Kinect, Leap Motion, or the like. In one or more embodiments, the input device includes a combination of devices, such as those disclosed herein.

[0104] Referring now to FIG. 1, one embodiment of a system 100 for providing access to quantum computers and quantum computer simulators is shown.

[0105] System 100 includes a digital computer 110. Digital computer 110 includes a processing unit 112 and a memory 114 that includes a computer program executable by the processing unit to, among other things, generate a request. As discussed above, it will be understood that digital computer 110 can be various types of digital computers. While an embodiment is disclosed in which the request is generated by digital computer 110, it will be understood that the request can be provided according to various alternative embodiments.

[0106] 1, system 100 further includes computing platform 120. Computing platform 120 includes at least one physics-inspired computer simulator that includes variable parameters and at least one optional physics-inspired computer. In one or more alternative embodiments, computing platform 120 includes at least one physics-inspired computer simulator.

[0107] The system 100 further includes a computing system 118. More precisely, the computing system 118 includes a communication interface 140, a control unit 126, an optional training unit 128, and a memory 130. The digital computer 110 is operatively connected to the computing system 118 via the communication interface 140 and using a data network, not shown.

[0108] The at least one optional physics-inspired computer of the computing platform can be of various types. For example, in one or more embodiments, the at least one optional physics-inspired computer includes a quantum computer 122. Those skilled in the art will appreciate that quantum computer 122 is an embodiment of a non-classical computer.

[0109] It will be appreciated that quantum computer 122 may be a variety of types of quantum computers, and indeed, in one or more embodiments, quantum computer 122 is selected from the group consisting of NISQ devices, superconducting quantum computers, ion trap quantum computers, quantum annealing, optical quantum computers, spin-based quantum dot computers, and photonic-based quantum computers.

[0110] In one or more embodiments, the at least one physics-inspired computer simulator type corresponds to an optional physics-inspired computer.

[0111] In one or more embodiments, the at least one physics-inspired computer simulator includes a quantum computer simulator 124. It will be understood by those skilled in the art that in one or more embodiments, the physics-inspired computer simulator is pre-trained.

[0112] In one or more embodiments, at least one quantum computer simulator 124 is represented by a probabilistic framework. Reinforcement learning is used by the optional training unit 128 to improve parameters defining the generative machine learning model. In one or more embodiments, the generative machine learning model is a restricted Boltzmann machine. In another embodiment, the generative machine learning model is selected from the group consisting of a deep Boltzmann machine, a forward propagation neural network, and a recurrent neural network. The network weights can be complex-valued, providing a complete description of both the amplitude and phase of the wave function. In one or more embodiments, the neural network parameters are optimized using static variational Monte Carlo sampling. In another embodiment, dynamic properties are of interest, where the neural network parameters are optimized using time-dependent variational Monte Carlo.

[0113] In one or more alternative embodiments, the quantum computer simulator is represented by a deep Boltzmann machine. In such an embodiment, two layers of hidden neurons are used to convey quantum correlations between the physics degrees of freedom in the visible layer. It will be appreciated that such a method reproduces the exact imaginary-time Hamiltonian expansion and is deterministic. Stochastic optimization of network parameters is not required to obtain a network representation of the ground state. It will further be appreciated that physical quantities can be measured by sampling configurations of both the physics degrees of freedom and the neuronal degrees of freedom.

[0114] In one or more alternative embodiments, the quantum computer simulator includes a recurrent neural network. More precisely, the quantum computer simulator includes a structure consisting of several stacked gated recurrent units, or GRUs. Using this type of scalable machine learning procedure allows for the reconstruction of both pure and mixed states. The method involves measurements of the quantum system. The learning procedure includes built-in approximate proofs of reconstruction and makes no assumptions about the purity of the state under examination. The learning procedure involves reducing state tomography to an unsupervised learning problem of the statistics of quantum measurements.

[0115] 1 , it will be appreciated that computing system 118 includes a communications interface 140 for receiving the request. It will be appreciated that the request is provided by a processing device. In one or more embodiments, the processing device is a digital computer. It will be further appreciated that the request may be provided according to various embodiments. In one or more embodiments, the request is provided by the processing device to communications interface 140 over a data network.

[0116] It will be appreciated that communication interface 140 may be implemented according to various embodiments. In one or more embodiments, communication interface 140 is implemented using an application programming interface (API) gateway configured to allow users to submit computational tasks and receive computational solutions.

[0117] In one or more embodiments, the application programming interface (API) gateway is programmed or configured to authenticate users of the computing system. In one of several embodiments, the application programming interface (API) gateway is programmed or configured to monitor system and data security. As an example, the application programming interface (API) gateway may use secure sockets layer (SSL) for encryption of requests and responses. In one of several embodiments, the application programming interface (API) gateway is programmed or configured to monitor data traffic.

[0118] The request includes at least one computational task for processing using at least one physics-inspired computer simulator.

[0119] It will be appreciated that in one or more embodiments, the request further includes an indication of a selection of at least one of the at least one optional physics-inspired computer and the at least one physics-inspired computer simulator.

[0120] It will be appreciated that the at least one computational task may be a variety of types of computational task. In one or more embodiments, the at least one computational task includes an optimization task. In one or more alternative embodiments, the at least one computational task includes sampling from a probability distribution. In one or more alternative embodiments, the at least one computational task includes any computational task selected from the group consisting of a database search, solving a Deutsch-Josa problem, solving a quantum chemistry related problem, and factoring an integer.

[0121] It will be further appreciated that in one or more embodiments, the request can further include an indication that the at least one computational task and at least one corresponding solution can be used for training purposes to improve the at least one physics-inspired computer simulator.

[0122] 1 , the computing system 118 further includes a memory 130. The memory 130 is operatively connected to the control unit 126 and the at least one physics-inspired computer simulator. The memory 130 is configured to store one or more of the at least one computational task. The memory 130 is further used to store a data set, which is included in a request and required to perform the at least one computational task. The memory 130 is further used to store variable parameters of the at least one physics-inspired computer simulator. Finally, the memory 130 is also used to store at least one corresponding solution received.

[0123] In one or more embodiments, the request further includes an indication of a selection of at least one of at least one optional physics-inspired computer and at least one physics-inspired computer simulator for performing the computational task, and memory 130 is further used to store the selection.

[0124] It will be appreciated that memory 130 is operatively connected to control unit 126 and at least one physics-inspired computer simulator. In one or more embodiments, memory 130 is accessed by control unit 126, optional training unit 128, and at least one physics-inspired computer simulator.

[0125] It will further be appreciated that memory 130 can be various types of memory. For example, in one or more embodiments, memory 130 comprises a database. Those skilled in the art will appreciate that many types of databases may be suitable for storing and retrieving data. In one or more embodiments, suitable databases include, by way of non-limiting example, relational databases, non-relational databases, object-oriented databases, object databases, entity-relationship model databases, associative databases, and XML databases. In one or more embodiments, the database is internet-based. In one or more embodiments, the database is web-based. In one or more embodiments, the database is cloud computing (e.g., on the cloud). In other embodiments, the database is based on one or more local computer storage devices. In one or more embodiments, a solution to the solved problem is maintained by the database. In one or more embodiments, data sent with the request is also stored in the database.

[0126] 1, it will be understood that the computing system 118 further includes a control unit 126. The control unit 126 is operatively connected to a communication interface 140, a memory 130, an optional training unit 128, and the computing platform 120.

[0127] The control unit 126 is used to convert the at least one computational task included in the received request into instructions for at least one of the at least one optional physics-inspired computer and the at least one physics-inspired computer simulator. It will be further understood that the control unit 126 is further used to transmit instructions to at least one of the at least one optional physics-inspired computer and the at least one physics-inspired computer simulator to perform the computational task.

[0128] The control unit 126 is further used to receive at least one corresponding solution to the computational task from at least one of the at least one optional physics-inspired computer and the at least one physics-inspired computer simulator.

[0129] It will be appreciated that in one or more embodiments, the computing system 118 further includes a training unit 128, which is operably connected to at least one physics-inspired computer simulator.

[0130] The optional training unit 128 is used to train the at least one physics-inspired computer simulator. It will be appreciated that in such an embodiment, if an indication is received that the at least one computational task and at least one corresponding solution are available for training purposes to improve the at least one physics-inspired computer simulator, the control unit 126 is further used to transmit instructions and the corresponding at least one solution to the optional training unit 128.

[0131] In one or more embodiments, the at least one physics-inspired computer simulator type corresponds to a physics-inspired computer, and the training unit 128 is used to train the at least one physics-inspired computer simulator using at least the instructions transmitted to the physics-inspired computer and using at least one corresponding solution obtained from the physics-inspired computer. It will be understood that in one or more embodiments, the instructions for the physics-inspired computer and the instructions for the at least one physics-inspired computer simulator are identical.

[0132] It will be further appreciated that, in one or more embodiments, at least one physics-inspired computer simulator includes a computer-implemented method that includes mimicking the output of the physics-inspired computer for a given input and using a training unit 128 to update at least one of the aforementioned variable parameters, thereby improving the corresponding performance.

[0133] It will be appreciated that optional training unit 128 may be various types of training units, and indeed, in one or more embodiments, optional training unit 128 used to improve the performance of at least one physics-inspired computer simulator is a computer-implemented method that uses artificial intelligence-based methods to update variable parameters of the physics-inspired computer simulator.

[0134] In one or more embodiments, the optional training unit 128 includes a neural network. Those skilled in the art will appreciate that the neural network can be various types of neural networks. In one or more embodiments, the neural network includes a restricted Boltzmann machine. In one or more other embodiments, the neural network includes a deep Boltzmann machine. In yet one or more other embodiments, the neural network includes a recurrent neural network. In an alternative embodiment, the neural network includes a feedforward neural network. It will be appreciated that the variable parameters can include weights of the neural network. In one or more embodiments, the training unit 128 includes a function approximator. The variable parameters can include function approximation parameters.

[0135] Those skilled in the art will understand that the training unit 128 may be implemented according to various embodiments. More precisely, the training unit 128 may include at least one of a tensor processing unit (TPU), a graphical processing unit (GPU), a field programmable gate array (FPGA), and an application specific integrated circuit (ASIC). Those skilled in the art will understand that various alternative embodiments may be provided for implementing the training unit 128.

[0136] In one or more embodiments, the control unit 126 is programmed to generate workers that perform the conversion of the computational task into instructions for at least one of the optional physics-inspired computer and the physics-inspired computer simulator. In such embodiments, the generated workers transmit instructions to at least one of the at least one optional physics-inspired computer and the at least one physics-inspired computer simulator according to the selection for performing the computational task. Furthermore, in this embodiment, the generated workers receive at least one corresponding solution and transmit the instructions and the at least one corresponding solution to the optional training unit 128.

[0137] It will be appreciated that in one or more embodiments, computing system 118 may further include an optional queuing unit, also referred to as a central queue, not shown in FIG. 1. Such optional queuing unit is programmed to queue multiple received requests according to criteria. It will be appreciated that in one or more embodiments, a given request is placed in a queue to maintain the order of requests in the queue and to prevent lost messages.

[0138] 2, shown is one embodiment of a system 2000. System 2000 includes digital computer 110, computing platform 120, and another embodiment of computing system 200 that is used to provide digital computer 110 with access to computing platform 120 that includes at least one of a quantum computer and a quantum computer simulator.

[0139] It will be appreciated that quantum computer 122 is an embodiment of a physics-inspired computer, while quantum computer simulator 124 is an embodiment of a physics-inspired computer simulator.

[0140] In this embodiment, it will be appreciated that the computing system 200 includes an API gateway 202 , a memory 130 , a control unit 126 , and a training unit 128 .

[0141] Additionally, as will be further described, the control unit 126 includes a central queue 204 , a cluster manager 208 , a worker farm 206 , and a central log 210 .

[0142] In this embodiment, a request including a computational task is sent by digital computer 110 to API gateway 202. The received request is then first processed by central queue 204 of control unit 126. Central queue 204 places the received request in a queue. Those skilled in the art will understand that the queue can have various sizes depending on the number of requests received.

[0143] The memory 130 is operatively connected to the central queue 204. In particular, the memory 130 communicates with, among other things, the central queue 204 to record queue states and transactions.

[0144] The central queue 204 also transmits the current state of the queue to the cluster manager 208 .

[0145] It will be appreciated that the cluster manager 208 initiates and controls the lifetime of a particular type of computational component. More precisely, the cluster manager 208 initiates at least one worker in the worker farm 206 to perform a computation, for example, translating received instructions for a particular quantum computation and controlling the digital processor and quantum processor to perform the computational task.

[0146] 2, the worker farm 206 includes a first worker 212, a second worker 214, and a third worker 216. Those skilled in the art will appreciate that a plurality of workers may include any number of workers. It will be appreciated that in one or more embodiments, workers that have completed their assigned tasks are then discarded by the cluster manager 208.

[0147] 2, the central log 210 communicates with the worker farm 206 to record all events. In practice, it will be appreciated that the central log 210 is responsible for tracking events that occur in different task executions. Thus, in one or more embodiments, every execution sends a corresponding log of events to the central log 210. The events are selected from the group consisting of the start of a task execution, the end of a task execution, an error that occurs during the execution of a task, a communication interruption, etc. Those skilled in the art will appreciate that various alternative embodiments may be provided for the events.

[0148] Although a particular embodiment of a computing system 200 is disclosed in FIG. 2, those skilled in the art will appreciate that various alternative embodiments of the computing system 200 are possible.

[0149] 3, an embodiment of a method for training a physics-inspired computer simulator using optional training units is shown. It will be appreciated that the purpose of training a physics-inspired computer simulator is to improve its performance.

[0150] More precisely, in one or more embodiments, a method for training a physics-inspired computer simulator includes using a machine learning training procedure.

[0151] It will be appreciated that the physics-inspired computer simulator may be any suitable physics-inspired computer simulator, for example, any of the physics-inspired computer simulators described herein with respect to system 100 disclosed in FIG. 1 or system 2000 disclosed in FIG. 2.

[0152] The optional training unit may be any suitable training unit, such as any training unit described herein with respect to computing system 118 disclosed in FIG. 1 or training unit 128 described in computing system 126 of FIG. 2. If a selection is made to use the at least one optional physics-inspired computer and approval is granted to use the computational task for training purposes, the at least one computational task and at least one corresponding solution may be provided to the optional training unit for purposes of training the at least one physics-inspired computer simulator. In response, the optional training unit may execute a procedure based on a machine learning protocol to update variable parameters of the at least one physics-inspired computer simulator. In one or more embodiments, if a selection is made to use the at least one optional physics-inspired computer and approval is granted to use the at least one computational task for training purposes, the instructions and the at least one corresponding solution are stored in memory.

[0153] 3, according to process step 302, at least one instance of an instruction and at least one corresponding solution are selected from memory using an optional training unit. Those skilled in the art will appreciate that the selection can be based on a variety of criteria, such as, for example, a time sequence of the stored instances of the instruction and the at least one corresponding solution. In one or more embodiments, the selection is based on at least one priority associated with the stored instances of the instruction and the at least one corresponding solution.

[0154] 3 , according to process step 304, a total error corresponding to the selected at least one instance of the instruction and at least one corresponding solution is calculated. Those skilled in the art will appreciate that various embodiments may be used to calculate the total error. In one or more embodiments, the calculation of the total error is based on mean squared error. In one or more alternative embodiments, the calculation of the total error is based on cross-entropy. In one or more alternative embodiments, the calculation of the total error is based on mean absolute error. It will further be appreciated that the calculation of the total error may depend on the machine learning procedure used to train the physics-inspired computer simulator.

[0155] Still referring to FIG. 3 , according to process step 306, a procedure based on a machine learning protocol is executed using optional training units. It will be appreciated that the machine learning protocol may be various types of machine learning protocols. In one or more embodiments, the machine learning protocol is a member selected from the group consisting of supervised learning, unsupervised learning, and reinforcement learning. In one or more embodiments, the procedure includes backpropagation. It will be appreciated that the procedure may include calculating derivatives with respect to variable parameters. The procedure may further include at least one member selected from the group consisting of batch gradient descent, stochastic gradient descent, and mini-batch gradient descent. It will be appreciated by those skilled in the art that the procedure may include any optimization method.

[0156] 3, according to process step 308, variable parameters of the physics-inspired simulator are updated using an optional training unit. It will be appreciated that in one or more embodiments, the variable parameters include weights of a neural network. It will be further appreciated that the neural network may include at least one member selected from the group consisting of a restricted Boltzmann machine, a deep Boltzmann machine, a feedforward neural network, and a recurrent neural network.

[0157] The updating can be performed according to various embodiments. For example, in one or more embodiments, the optional training unit directly updates the variable parameters of the physics-inspired computer simulator. In one or more other embodiments, the optional training unit updates the variable parameters of the physics-inspired computer simulator stored in memory. The physics-inspired computer simulator then accesses and reads the updated values ​​stored in memory. Those skilled in the art will appreciate that various alternative embodiments may be provided for updating the variable parameters of the physics-inspired computer simulator.

[0158] Referring now to FIG. 4 , an embodiment of a method for enabling remote access to a physics-inspired computing resource and a simulation thereof is shown. In one or more embodiments, the physics-inspired computing resource includes at least one optional physics-inspired computer. It will be understood that the optional physics-inspired computer can be any suitable physics-inspired computer, for example, any of the physics-inspired computers described herein with respect to system 100 disclosed in FIG. 1 or system 2000 disclosed in FIG. 2. In one or more embodiments, the optional physics-inspired computer includes a quantum computer. The quantum computer can be any suitable quantum computer, for example, any of the quantum computers described herein with respect to system 100 disclosed in FIG. 1 or system 2000 disclosed in FIG. 2.

[0159] Thus, while quantum computers and quantum computer simulators are further disclosed in Figure 4, it should be understood that the method disclosed in Figure 4 more generally enables remote access to physics-inspired computers, an example of which is quantum computers and physics-inspired computer simulators, an example of which is a quantum computer simulator.

[0160] Still referring to FIG. 4 , according to processing step 402, a request is received, the request including at least one computational task and an indication of a selection of at least one of a physics-inspired computer and at least one physics-inspired computer simulator to use to process the at least one computational task.

[0161] It will be appreciated that the request may be received according to various embodiments. In one or more embodiments, the request is received by a communications interface of a computing system over a data network. In one or more embodiments, the request is transmitted by a digital computer operably connected to the communications interface over a data network.

[0162] It will be appreciated that the digital computer may be any suitable digital computer, for example, any digital computer described herein with respect to system 100 disclosed in FIG. 1 or system 2000 disclosed in FIG. 2.

[0163] It will further be understood that the communication interface may be various types of communication interfaces. In one or more embodiments, the communication interface includes an API gateway. The selection is of at least one of the at least one optional physics-inspired computer and the at least one physics-inspired computer simulator. It will be understood that in one or more embodiments, the request further includes an indication that the at least one corresponding generated solution to the at least one corresponding computational task is usable for training purposes. More precisely, as described further below, the at least one corresponding generated solution and the at least one corresponding computational task can then be used to train the at least one physics-inspired computer simulator.

[0164] Those skilled in the art will appreciate that the at least one computational task can be a variety of types of computational task. In one or more embodiments, the at least one computational task includes an optimization task. In one or more alternative embodiments, the at least one computational task includes sampling from a probability distribution. In one or more alternative embodiments, the at least one computational task includes any member of the group consisting of a database search, solving a Deutsch-Josa problem, solving a quantum chemistry related problem, and factoring an integer.

[0165] 4, according to process step 404, the received request is placed in a queue. In one or more embodiments, the received request is placed in the queue using a queuing unit located in the computing system. It will be understood that this process step is optional.

[0166] 4, according to processing step 406, the received request is converted. It will be understood that the received request may or may not be placed in a queue depending on whether processing step 404 is performed or not. In practice, it will be understood that the received request is converted into instructions suitable for at least one of a quantum computer and a quantum computer simulator.

[0167] In one or more embodiments, the conversion is performed using a control unit of the computing system. It will be appreciated that various alternative embodiments may be provided for converting the received request. It will be appreciated that the quantum computer simulator may be any suitable quantum computer simulator, for example, any quantum computer simulator described herein with respect to system 100 disclosed in FIG. 1 or system 2000 disclosed in FIG. 2.

[0168] Continuing with reference to FIG. 4 , according to process step 408, a determination is made. It will be understood that the purpose of the determination is to determine whether a quantum computer or quantum computer simulator should be used to execute the instruction. It will be understood that the determination may be made according to various embodiments. In one or more embodiments, the determination is made based on a user selection. In one or more alternative embodiments, the determination is made based on various other considerations.

[0169] If the determination is that a quantum computer should be used to execute the instructions, the instructions are transmitted to the quantum computer, according to process step 410. It will be appreciated that the instructions may be transmitted to the quantum computer according to various embodiments known to those skilled in the art.

[0170] According to process step 414, the instructions are executed using the quantum computer to generate at least one corresponding solution resulting from the execution of the instructions.

[0171] If the determination is that a quantum computer simulator should be used to execute the instructions, the instructions are transmitted to the quantum computer simulator, according to process step 412. It will be appreciated that the instructions may be transmitted to the quantum computer simulator according to various embodiments known to those skilled in the art.

[0172] According to process step 416, the instructions are then executed using a quantum computer simulator to generate at least one corresponding solution resulting from the execution of the instructions.

[0173] 4, according to process step 418, at least one corresponding generated solution resulting from the execution of the instructions is received. In one or more embodiments, the at least one corresponding generated solution is received using a control unit of the computing system.

[0174] According to process step 420, the at least one corresponding generated solution is stored in memory. In one or more embodiments, the at least one corresponding generated solution is stored in memory of the computing system by the control unit.

[0175] 4, according to process step 422, at least one corresponding generated solution is provided to a digital computer. In one or more embodiments, at least one corresponding generated solution is provided to the digital computer via a data network.

[0176] It will be appreciated that processing steps 420 and 422 are embodiments that provide at least one corresponding generated solution.

[0177] It will be appreciated that in embodiments in which the at least one corresponding generated solution is obtained from at least one quantum computer, the method may further include training a quantum computer simulator using the at least one corresponding generated solution and the at least one computational task.

[0178] It will further be appreciated that in one or more embodiments, training is performed if the request includes an indication that at least one computational task and at least one corresponding generated solution is usable for training purposes.

[0179] In one or more embodiments, training includes executing a procedure based on a machine learning protocol using at least one corresponding generated solution and at least one computational task, and updating variable parameters of a quantum computer simulator accordingly.

[0180] 5, an embodiment of a flowchart illustrating an embodiment of a method for enabling remote access to a computing platform including at least one physics-inspired computer simulator including variable parameters is shown. More precisely, FIG. 5 describes a method for using the physics-inspired computer simulator over a data network.

[0181] It will be appreciated that the physics-inspired computer simulator can be any suitable physics-inspired computer, such as any of the physics-inspired computers described herein with respect to system 100 disclosed in Figure 1 or system 2000 disclosed in Figure 2. In one or more embodiments, parameters of the physics-inspired computer simulator are stored in memory.

[0182] According to process step 502, a request including at least one computational task is received.

[0183] It will be appreciated that the request may be received according to various embodiments. In one or more embodiments, the request is received by a communications interface of a computing system over a data network. In one or more embodiments, the request is transmitted by a digital computer operably connected to the communications interface over the data network.

[0184] It should be understood that the digital computer may be any suitable digital computer, for example, any digital computer described herein with respect to system 100 disclosed in FIG. 1 or system 2000 disclosed in FIG. 2.

[0185] 5, according to process step 504, the received request is translated. In practice, it will be appreciated that the received request is translated into instructions suitable for at least one physics-inspired computer simulator.

[0186] In one or more embodiments, the conversion is performed using a control unit of the computing system. It will be appreciated that various alternative embodiments may be provided for converting the received request.

[0187] Still referring to FIG. 5, according to process step 506, the instructions are transmitted to at least one physics-inspired computer simulator.

[0188] In one or more embodiments, the instructions are transmitted to at least one physics-inspired computer simulator using a control unit. In one or more embodiments, the instructions are stored in a memory.

[0189] Still referring to FIG. 5, according to process step 508, the instructions are executed using at least one physics-inspired computer simulator, and at least one corresponding solution is generated by the physics-inspired computer simulator.

[0190] Still referring to FIG. 5, according to process step 510, at least one corresponding generated solution is received.

[0191] It will be appreciated that the at least one corresponding generated solution may be received according to various embodiments. In one or more embodiments, the at least one corresponding generated solution is received using a control unit of a computing system. In one or more embodiments, the at least one corresponding generated solution is stored in a memory.

[0192] Still referring to FIG. 5, according to process step 512, at least one corresponding generated solution is provided to a digital computer.

[0193] It will be appreciated that the at least one corresponding generated solution may be provided to a digital computer according to various embodiments, in one or more embodiments, the at least one corresponding generated solution is provided to the digital computer using a communications interface of the computing system and a data network.

[0194] It will be appreciated that one or more embodiments of the method and computing system disclosed herein are highly advantageous for a variety of reasons.

[0195] More precisely, an advantage of one or more embodiments of the methods and computing systems disclosed herein is that they enable access to physics-inspired computer simulators trained using real computational tasks, such as simulators of quantum devices, which are relatively cheaper than access to quantum devices.

[0196] Another advantage of one or more embodiments of the methods and computing systems disclosed herein is that they allow for the emulation of physics-inspired computers such as quantum computers.

[0197] Another advantage of one or more embodiments of the methods and computing systems disclosed herein is that they enable user-provided computational tasks to be used to improve a physics-inspired computer simulator.

[0198] Section 1 1. A computing system for enabling a processing device to remotely access a computing platform over a network, the computing platform including at least one physics-inspired computer simulator including variable parameters, the computing system comprising: a) a communications interface for receiving a request provided by a processing device, the request including at least one computational task for processing using at least one physics-inspired computer simulator, the computational task including variable parameters; b) a control unit operatively connected to the communication interface and to at least one physics-inspired computer simulator including variable parameters, the control unit for: converting a received request into instructions for the at least one physics-inspired computer simulator; transmitting the instructions to the at least one physics-inspired computer simulator to perform at least one computational task; and receiving at least one corresponding solution; c) a memory operatively connected to the control unit and the at least one physics-inspired computer simulator, the memory being for storing one or more of the at least one computational task, a data set included in the received request, variable parameters of the at least one physics-inspired computer simulator, and the at least one received corresponding solution; a computing system including:

[0199] Section 2 10. The computing system of claim 1, the computing platform further includes at least one physics-inspired computer; Additionally, the request further includes selecting at least one of the at least one physics-inspired computer simulator and the at least one physics-inspired computer; Additionally, the control unit is further operably connected to at least one physics-inspired computer; Furthermore, if the selection of the request is a selection for the at least one physics-inspired computer, the control unit is further used for converting the received request into instructions for the at least one physics-inspired computer and transmitting the instructions to the at least one physics-inspired computer to perform the at least one computational task; Additionally, the control unit is further used to receive at least one corresponding solution from the at least one physics-inspired computer. Computing system.

[0200] Section 3 10. The computing system of claim 1, further comprising a training unit operably connected to the at least one physics-inspired computer simulator, the training unit being for training the at least one physics-inspired computer simulator.

[0201] Section 4 3. The computing system of claim 2, wherein the computing system includes a training unit operatively connected to at least one physics-inspired computer simulator, the training unit being for training the at least one physics-inspired computer simulator.

[0202] Section 5 5. The computing system of claim 4, at least one physics-inspired computer simulator type corresponds to a physics-inspired computer; Further, the training unit is used to train at least one physics-inspired computer simulator using at least the instructions transmitted to the physics-inspired computer and using at least one corresponding solution obtained from the physics-inspired computer. Computing system.

[0203] Section 6 10. The computing system of claim 1, wherein the communication interface receives a plurality of requests, and the computing platform further includes a queuing unit for queuing the received plurality of requests according to a criterion.

[0204] Section 7 10. The computing system of claim 5, wherein if the corresponding request includes an indication that the at least one computational task and at least one corresponding solution are usable for training purposes, the training unit trains the at least one physics-inspired computer simulator using the corresponding instructions transmitted to the physics-inspired computer and the at least one corresponding solution obtained from the physics-inspired computer.

[0205] Section 8 10. The computing system of claim 1, wherein the processing device for remotely accessing the computing system includes a digital computer operably connected to the communication interface via a data network.

[0206] Section 9 3. The computing system of claim 2, wherein the instructions for the physics-inspired computer and the instructions for the at least one physics-inspired computer simulator are identical.

[0207] Section 10 10. The computing system of claim 1, wherein the computing platform comprises a distributed computing system.

[0208] Section 11 3. The computing system of claim 2, wherein the physics-inspired computer comprises a non-classical computer.

[0209] Section 12 12. The computing system of claim 11, wherein the non-classical computer is selected from the group consisting of a NISQ device, a quantum computer, a superconducting quantum computer, an ion trap quantum computer, quantum annealing, an optical quantum computer, a spin-based quantum dot computer, and a photonic-based quantum computer.

[0210] Section 13 10. The computing system of claim 1, wherein the at least one physics-inspired computer simulator includes a computer-implemented method, the method comprising: a) mimicking the output of a physics-inspired computer for a given input; and b) using the training unit to update at least one of the variable parameters, thereby improving the corresponding performance; a computing system including:

[0211] Section 14 5. The computing system of any one of claims 3 to 4, wherein the training unit is selected from the group consisting of a tensor processing unit (TPU), a graphical processing unit (GPU), a field programmable gate array (FPGA), and an application specific integrated circuit (ASIC).

[0212] Section 15 10. The computing system of claim 1, wherein the at least one physics-inspired computer simulator includes a neural network.

[0213] Section 16 1. A computer-implemented method for enabling remote access to a computing platform, the computing platform including at least one physics-inspired computer simulator including variable parameters, the method comprising: a) receiving a request at a communication interface, the request including at least one computational task for processing using at least one physics-inspired computer simulator; b) converting at least one computational task of the received request into instructions suitable for at least one physics-inspired computer simulator; c) providing instructions to at least one physics-inspired computer simulator; d) receiving at least one corresponding generated solution resulting from the execution of the instructions; and e) providing at least one corresponding generated solution; and 11. A computer-implemented method comprising:

[0214] Section 17 17. The computer-implemented method of claim 16, the computing system further includes at least one physics-inspired computer and a training unit for training the at least one physics-inspired computer simulator; Additionally, the received request further includes an indication of a selection of at least one of the physics-inspired computer and the at least one physics-inspired computer simulator for use in processing the at least one computational task; Further, at least one computational task of the request is translated into instructions suitable for at least one of a physics-inspired computer and at least one physics-inspired computer simulator; Further, providing the instructions is performed in at least one of a physics-inspired computer and at least one physics-inspired computer simulator in response to the indication of the selection. Computer-implemented methods.

[0215] Section 18 17. The computer-implemented method of claim 16, wherein the request is received from a digital computer operably connected to the communication interface using a data network, and further wherein at least one corresponding generated solution is provided to the digital computer.

[0216] Section 19 18. The computer-implemented method of claim 17, at least one corresponding generated solution is obtained from at least one physics-inspired computer; further comprising training at least one physics-inspired computer simulator using the at least one corresponding generated solution and the at least one computational task. Computer-implemented methods.

[0217] Section 20 20. The computer-implemented method of claim 19, wherein training is performed if the request includes an indication that at least one computational task and at least one corresponding generated solution are usable for training purposes.

[0218] Section 21 20. The computer-implemented method of claim 19, wherein the training comprises: a) performing a procedure based on a machine learning protocol using at least one corresponding generated solution and at least one computational task; b) updating variable parameters of a physics-inspired computer simulator accordingly; and 11. A computer-implemented method comprising:

[0219] Section 22 Clause 17. The computer-implemented method of clause 16, further comprising storing the instructions and at least one corresponding generated solution.

Claims

1. 1. A computing system for enabling a processing device to remotely access a computing platform over a network, the computing platform including at least one physics-inspired computer simulator including variable parameters, the computing system comprising: a) at least one physics-inspired computer simulator, the physics-inspired computer simulator including a set of variable parameters, that emulates a physics-inspired computer output for a given input, the emulation not including a simulation of a physics-inspired computer computation; b) a communications interface for receiving a request provided by a processing device, said request including at least one computational task for processing using said at least one physics-inspired computer simulator; and c) a control unit operatively connected to the communication interface and to the at least one physics-inspired computer simulator including variable parameters, the control unit for: converting the received request into instructions for the at least one physics-inspired computer simulator; transmitting the instructions to the at least one physics-inspired computer simulator to perform the at least one computational task; and receiving at least one corresponding solution; d) a memory operatively connected to the control unit and the at least one physics-inspired computer simulator, the memory being for storing one or more of the at least one computational task, a data set included in the received request, the variable parameters of the at least one physics-inspired computer simulator, and the received at least one corresponding solution; a computing system including:

2. 10. The computing system of claim 1, the computing platform further comprises at least one physics-inspired computer; Additionally, the request further includes a selection of at least one of the at least one physics-inspired computer simulator and the at least one physics-inspired computer; Additionally, the control unit is further operably connected to the at least one physics-inspired computer; Furthermore, if the selection of the request is a selection for the at least one physics-inspired computer, the control unit is further used for converting the received request into instructions for the at least one physics-inspired computer and transmitting the instructions to the at least one physics-inspired computer to perform the at least one computational task; and wherein the control unit is further adapted to receive at least one corresponding solution from the at least one physics-inspired computer. Computing system.

3. 10. The computing system of claim 1, further comprising a training unit operatively connected to the at least one physics-inspired computer simulator, the training unit being for training the at least one physics-inspired computer simulator.

4. 3. The computing system of claim 2, further comprising a training unit operatively connected to the at least one physics-inspired computer simulator, the training unit being for training the at least one physics-inspired computer simulator.

5. 5. The computing system of claim 4, a type of the at least one physics-inspired computer simulator corresponding to the physics-inspired computer; further, the training unit is used to train the at least one physics-inspired computer simulator using at least the instructions transmitted to the physics-inspired computer and using at least one corresponding solution obtained from the physics-inspired computer. Computing system.

6. 10. The computing system of claim 1, wherein the communication interface receives a plurality of requests, and wherein the computing platform further comprises a queuing unit for queuing the received plurality of requests according to a criterion.

7. 6. The computing system of claim 5, wherein if a corresponding request includes an indication that the at least one computational task and the at least one corresponding solution are usable for training purposes, the training unit trains the at least one physics-inspired computer simulator using the corresponding instructions transmitted to the physics-inspired computer and the at least one corresponding solution obtained from the physics-inspired computer.

8. 2. The computing system of claim 1, wherein the processing device for remotely accessing the computing system comprises a digital computer operably connected to the communication interface via a data network.

9. 3. The computing system of claim 2, wherein the instructions for the physics-inspired computer and the instructions for the at least one physics-inspired computer simulator are identical.

10. 10. The computing system of claim 1, wherein the computing platform comprises a distributed computing system.

11. 3. The computing system of claim 2, wherein the physics-inspired computer comprises a non-classical computer.

12. 12. The computing system of claim 11, wherein the non-classical computer is selected from the group consisting of a NISQ device, a quantum computer, a superconducting quantum computer, an ion trap quantum computer, quantum annealing, an optical quantum computer, a spin-based quantum dot computer, and a photonic-based quantum computer.

13. 4. The computing system of claim 3, wherein the at least one physics-inspired computer simulator comprises a computer-implemented method, the method comprising: a) mimicking the output of a physics-inspired computer for a given input; b) using the training unit to update at least one of the variable parameters, thereby improving corresponding performance; a computing system including:

14. 5. The computing system of claim 3, wherein the training unit is selected from the group consisting of a tensor processing unit (TPU), a graphical processing unit (GPU), a field programmable gate array (FPGA), and an application specific integrated circuit (ASIC).

15. 10. The computing system of claim 1, wherein the at least one physics-inspired computer simulator comprises a neural network.

16. 1. A computer-implemented method for enabling remote access to a computing platform, the computing platform including at least one physics-inspired computer simulator including variable parameters, the method comprising: a) receiving a request at a communication interface, the request including at least one computational task for processing using at least one physics-inspired computer simulator; b) converting the at least one computational task of the received request into instructions suitable for the at least one physics-inspired computer simulator; c) providing said instructions to said at least one physics-inspired computer simulator; and d) receiving at least one corresponding generated solution that mimics a physics-inspired computer output resulting from execution of the instructions, wherein the mimicry does not include a simulation of a physics-inspired computer calculation; and e) providing said at least one corresponding generated solution; and 11. A computer-implemented method comprising:

17. 17. The computer-implemented method of claim 16, the computing platform further includes at least one physics-inspired computer and a training unit for training the at least one physics-inspired computer simulator; and wherein the received request further includes an indication of a selection of at least one of the physics-inspired computer and the at least one physics-inspired computer simulator for use in processing the at least one computational task; further comprising: translating the at least one computational task of the request into instructions suitable for at least one of the physics-inspired computer and the at least one physics-inspired computer simulator; further, the providing of the instructions is performed in at least one of the physics-inspired computer and the at least one physics-inspired computer simulator in response to the indication of selection. Computer-implemented methods.

18. 17. The computer-implemented method of claim 16, wherein the request is received from a digital computer operatively connected to the communication interface using a data network, and further wherein the at least one corresponding generated solution is provided to the digital computer.

19. 18. The computer-implemented method of claim 17, the at least one corresponding generated solution is obtained from the at least one physics-inspired computer; and further comprising training the at least one physics-inspired computer simulator using the at least one corresponding generated solution and the at least one computational task. Computer-implemented methods.

20. 20. The computer-implemented method of claim 19, wherein the training is performed if the request includes an indication that the at least one computational task and the at least one corresponding generated solution are usable for training purposes.

21. 20. The computer-implemented method of claim 19, wherein the training comprises: a) performing a procedure based on a machine learning protocol using the at least one corresponding generated solution and the at least one computational task; b) updating the variable parameters of the physics-inspired computer simulator accordingly; and 11. A computer-implemented method comprising:

22. 17. The computer-implemented method of claim 16, further comprising storing the instructions and the at least one corresponding generated solution.

Images