A combinatorial optimization problem processing device is for associating a combinatorial optimization problem having N elements with an Ising model to process the combinatorial optimization problem. The combinatorial optimization problem processing device includes: a 1×2 Mach-Zehnder optical modulator that receives a polarized clock pulse train; an optical interference circuit that receives polarized clock pulse trains that were modulated by the Mach-Zehnder optical modulator; an optical coupler that couples output of the optical interference circuit with an initialization optical pulse train that creates a neutral state with respect to interactions between the elements; and a modulation signal generator that performs waveform shaping on an electrical signal obtained by photoelectrically converting an output signal of the optical coupler, generates a modulation signal for the Mach-Zehnder optical modulator, and externally outputs a monitor signal that represents a solution to the optimization problem. The optical interference circuit repeatedly allows a predetermined interaction in the Ising model to occur from the neutral state at a period corresponding to the N pulses of the polarized clock pulse train.

A photonics reservoir computing system is described. The system is configured for propagating at least one optical signal so as to create resulting radiation signals in the output channels. The photonics reservoir computing system further comprises weighting elements for weighting signals from the output channels, and at least one optical detector for optically detecting signals from the output channels. The system is adapted for estimating signals from the output channels through an output of the optical detector.

A combinatorial optimization problem processing device is for associating a combinatorial optimization problem having N elements with an Ising model to process the combinatorial optimization problem. The combinatorial optimization problem processing device includes: a 1×2 Mach-Zehnder optical modulator that receives a polarized clock pulse train; an optical interference circuit that receives polarized clock pulse trains that were modulated by the Mach-Zehnder optical modulator; an optical coupler that couples output of the optical interference circuit with an initialization optical pulse train that creates a neutral state with respect to interactions between the elements; and a modulation signal generator that performs waveform shaping on an electrical signal obtained by photoelectrically converting an output signal of the optical coupler, generates a modulation signal for the Mach-Zehnder optical modulator, and externally outputs a monitor signal that represents a solution to the optimization problem. The optical interference circuit repeatedly allows a predetermined interaction in the Ising model to occur from the neutral state at a period corresponding to the N pulses of the polarized clock pulse train.

A passive photonics reservoir computing system comprises an optical waveguide based structure comprising a plurality of discrete nodes and a plurality of passive waveguide interconnections between the nodes for propagating the at least one photonic signal between the nodes, in which each discrete node is adapted for passively relaying the at least one photonic wave over the passive waveguide interconnections connected thereto, wherein the optical waveguide based structure comprises at least one multimode Y-junction configured for connecting three waveguides using a taper section wherein the taper section is not perfectly adiabatic. A training scheme uses a passive photonics computing system.

A photonics reservoir computing system is described. The system is configured for propagating at least one optical signal so as to create resulting radiation signals in the output channels. The photonics reservoir computing system further comprises weighting elements for weighting signals from the output channels, and at least one optical detector for optically detecting signals from the output channels. The system is adapted for estimating signals from the output channels through an output of the optical detector.

A passive photonics reservoir computing system comprises an optical waveguide based structure comprising a plurality of discrete nodes and a plurality of passive waveguide interconnections between the nodes for propagating the at least one photonic signal between the nodes, in which each discrete node is adapted for passively relaying the at least one photonic wave over the passive waveguide interconnections connected thereto, wherein the optical waveguide based structure comprises at least one multimode Y-junction configured for connecting three waveguides using a taper section wherein the taper section is not perfectly adiabatic. A training scheme uses a passive photonics computing system.

A photonic circuit for generating an ultrafast clock for a photonic processor. The photonic circuit includes a beam splitter, a phase shifter, a first photonic combiner coupled to the beam splitter and the phase shifter, and a second photonic combiner coupled to the beam splitter and the first photonic combiner. The beam splitter splits a received photonic seed clock signal into a first photonic seed clock signal and a second photonic seed clock signal. The phase shifter shifts a phase of a received photonic signal to generate a phase-shifted version of the photonic signal. The first photonic combiner combines the second photonic seed clock signal with the phase-shifted version of the photonic signal to generate a first combined photonic signal. The second photonic combiner combines a delayed and attenuated version of the first photonic seed clock signal with the first combined photonic signal to generate a photonic clock signal.