Implementation of a fast digital optical matrix-vector multiplier using a holographic look-up table and residue arithmetic

A small microring array that performs large complex-valued matrix-vector multiplication

As an important computing operation, photonic matrix-vector multiplication is widely used in photonic neutral networks and signal processing. However, conventional incoherent matrix-vector multiplication focuses on real-valued operations, which cannot work well in complex-valued neural networks and discrete Fourier transform. In this paper, we propose a systematic solution to extend the matrix computation of microring arrays from the real-valued field to the complex-valued field, and from small-scale (i.e., 4 × 4) to large-scale matrix computation (i.e., 16 × 16). Combining matrix decomposition and matrix partition, our photonic complex matrix-vector multiplier chip can support arbitrary large-scale and complex-valued matrix computation. We further demonstrate Walsh-Hardmard transform, discrete cosine transform, discrete Fourier transform, and image convolutional processing. Our scheme provides a path towards breaking the limits of complex-valued computing accelerator in conventional incoherent optical architecture. More importantly, our results reveal that an integrated photonic platform is of huge potential for large-scale, complex-valued, artificial intelligence computing and signal processing.

Photonic matrix multiplication lights up photonic accelerator and beyond

Matrix computation, as a fundamental building block of information processing in science and technology, contributes most of the computational overheads in modern signal processing and artificial intelligence algorithms. Photonic accelerators are designed to accelerate specific categories of computing in the optical domain, especially matrix multiplication, to address the growing demand for computing resources and capacity. Photonic matrix multiplication has much potential to expand the domain of telecommunication, and artificial intelligence benefiting from its superior performance. Recent research in photonic matrix multiplication has flourished and may provide opportunities to develop applications that are unachievable at present by conventional electronic processors. In this review, we first introduce the methods of photonic matrix multiplication, mainly including the plane light conversion method, Mach-Zehnder interferometer method and wavelength division multiplexing method. We also summarize the developmental milestones of photonic matrix multiplication and the related applications. Then, we review their detailed advances in applications to optical signal processing and artificial neural networks in recent years. Finally, we comment on the challenges and perspectives of photonic matrix multiplication and photonic acceleration.

A single shot coherent Ising machine based on a network of injection-locked multicore fiber lasers

Combinatorial optimization problems over large and complex systems have many applications in social networks, image processing, artificial intelligence, computational biology and a variety of other areas. Finding the optimized solution for such problems in general are usually in non-deterministic polynomial time (NP)-hard complexity class. Some NP-hard problems can be easily mapped to minimizing an Ising energy function. Here, we present an analog all-optical implementation of a coherent Ising machine (CIM) based on a network of injection-locked multicore fiber (MCF) lasers. The Zeeman terms and the mutual couplings appearing in the Ising Hamiltonians are implemented using spatial light modulators (SLMs). As a proof-of-principle, we demonstrate the use of optics to solve several Ising Hamiltonians for up to thirteen nodes. Overall, the average accuracy of the CIM to find the ground state energy was ~90% for 120 trials. The fundamental bottlenecks for the scalability and programmability of the presented CIM are discussed as well.

For specific computation problems where electronic digital processors have shortcomings in simulating, an analog optical system may be a solution, Here, the authors present an analog all-optical implementation of a coherent Ising machine based on a network of injection-locked multicore fiber lasers.

Optical implementation and applications of closest-vector selection in neural networks

Closest-vector selection is a process that underlies one technique for sending compressed-signal sets over noisy communication channels. Recently it has had application in radar target identification, speech and image analysis, pattern classification, and neural-net training. Various applications of closest-vector selection are discussed, and the design of an all-optical system that performs closest-vector selection is presented.

Acoustooptic linear heterodyned complex-valued matrix-vector processor

A new analog (linear) optical vector-inner-product and matrix-vector processor architecture is described. It employs linear input laser-diode arrays and a multichannel acoustooptic cell. Attention is given to the selection of the operating modes for all devices, error correction, the ability of the system to operate on complex-valued data, and its stability with respect to temperature, data history, and component variations. Laboratory data are provided verifying the potential of the system architecture to achieve 9-10-bit accuracy and complex-valued multiplication.

Optical parallel register transfer microoperations using holographic symbolic substitutions

Optical register transfer microoperations are proposed. Based on an optical holographic associative symbolic substitution, a hybrid optical word-parallel bit-serial register transfer processor architecture is described. Preliminary experimental results are included.

Reconfigurable bipolar analog optical crossbar switch

A differential technique is proposed for obtaining bipolar analog {-1 . . . 0 . . . 1} inputs, outputs, and connection weights in a mask-type optical crossbar switch. The technique compensates for the limitations of current component technology and permits the use of inexpensive, low-contrast spatial light modulators for reconfigurable switching matrices. The conclusions of this work may be extended to design improvements in a wide variety of optical computing implementations.