Methods and systems for designing image processing devices

    公开(公告)号:US11455443B1

    公开(公告)日:2022-09-27

    申请号:US16555419

    申请日:2019-08-29

    Abstract: A computer-implemented method for designing an image processing device includes defining a loss function within a simulation space composed of a plurality of voxels; defining an initial structure for one or more physical features of a metasurface and one or more architectural features of a neural network in the simulation space; determining, using a computer system, values for at least one structural parameter, and/or at least one functional parameter for the one or more physical features and at least one architectural parameter for the one or more architectural features, using a numerical solver to solve Maxwell's equations so that a loss determined according to the loss function is within a threshold loss; defining a final structure of the metasurface based on the values for the one or more structural parameters; and defining a final structure of the neural network based on the values for the at least one architectural parameter.

    TECHNIQUES OF ROBUST INVERSE DESIGN THAT ACCOUNT FOR MANUFACTURING VARIABILITIES DUE TO OPERATING CONDITIONS

    公开(公告)号:US20220077948A1

    公开(公告)日:2022-03-10

    申请号:US17529112

    申请日:2021-11-17

    Abstract: Embodiments of techniques for inverse design of physical devices are described herein, in the context of generating designs for photonic integrated circuits (including a multi-channel photonic demultiplexer). In some embodiments, an initial design of the physical device is received, and a plurality of sets of operating conditions for fabrication of the physical device are determined. In some embodiments, the performance of the physical device as fabricated under the sets of operating conditions is simulated, and a total performance loss value is backpropagated to determine a gradient to be used to update the initial design. In some embodiments, instead of simulating fabrication of the physical device under the sets of operating conditions, a robustness loss is determined and combined with the performance loss to determine the gradient.

    Techniques of robust inverse design that account for manufacturing variabilities due to operating conditions

    公开(公告)号:US11196503B2

    公开(公告)日:2021-12-07

    申请号:US16796660

    申请日:2020-02-20

    Abstract: Embodiments of techniques for inverse design of physical devices are described herein, in the context of generating designs for photonic integrated circuits (including a multi-channel photonic demultiplexer). In some embodiments, an initial design of the physical device is received, and a plurality of sets of operating conditions for fabrication of the physical device are determined. In some embodiments, the performance of the physical device as fabricated under the sets of operating conditions is simulated, and a total performance loss value is backpropagated to determine a gradient to be used to update the initial design. In some embodiments, instead of simulating fabrication of the physical device under the sets of operating conditions, a robustness loss is determined and combined with the performance loss to determine the gradient.

    Two-channel integrated photonic wavelength demultiplexer

    公开(公告)号:US11187854B2

    公开(公告)日:2021-11-30

    申请号:US16685473

    申请日:2019-11-15

    Abstract: A two-channel photonic demultiplexer includes an input region to receive a multi-channel optical signal, two output regions, each adapted to receive a corresponding one of two distinct wavelength channels demultiplexed from the multi-channel optical signal, and a dispersive region including a first material and a second material inhomogeneously interspersed to form a plurality of interfaces that collectively structure the dispersive region to optically separate each of the two distinct wavelength channels from the multi-channel optical signal and respectively guide the first distinct wavelength channel to a first output region and the second distinct wavelength channel to the second output region when the input region receives the multi-channel optical signal. At least one of the first material or the second material is structured within the dispersive region to be schematically reproducible by a feature shape with a pre-determined width.

    TECHNIQUES OF ROBUST INVERSE DESIGN THAT ACCOUNT FOR MANUFACTURING VARIABILITIES DUE TO OPERATING CONDITIONS

    公开(公告)号:US20210266088A1

    公开(公告)日:2021-08-26

    申请号:US16796660

    申请日:2020-02-20

    Abstract: Embodiments of techniques for inverse design of physical devices are described herein, in the context of generating designs for photonic integrated circuits (including a multi-channel photonic demultiplexer). In some embodiments, an initial design of the physical device is received, and a plurality of sets of operating conditions for fabrication of the physical device are determined. In some embodiments, the performance of the physical device as fabricated under the sets of operating conditions is simulated, and a total performance loss value is backpropagated to determine a gradient to be used to update the initial design. In some embodiments, instead of simulating fabrication of the physical device under the sets of operating conditions, a robustness loss is determined and combined with the performance loss to determine the gradient.

    Systems and methods for determining fabrication loss of segmented designs using paintbrush patterns

    公开(公告)号:US10885262B1

    公开(公告)日:2021-01-05

    申请号:US16796500

    申请日:2020-02-20

    Abstract: In some embodiments, a design verification system is provided that is configured to perform actions for ensuring fabricability of a segmented design. The design verification system searches a proposed segmented design for a paintbrush pattern to determine a positive paintbrush loss, and searches for an inverse paintbrush pattern to determine a negative paintbrush loss. The design verification system combines the positive paintbrush loss and the negative paintbrush loss to obtain a total paintbrush loss that indicates whether or not the proposed segmented design is fabricable. If the total paintbrush loss indicates that the proposed segmented design is not fabricable, the design verification system updates the proposed segmented design based on a gradient of the total paintbrush loss.

    Finite floating-point numerical simulation and optimization

    公开(公告)号:US11842129B1

    公开(公告)日:2023-12-12

    申请号:US16886602

    申请日:2020-05-28

    CPC classification number: G06F30/23 G06F7/483 G06N3/08 G06F2111/10

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for correcting finite floating-point numerical simulation and optimization. Defining a loss function within a simulation space composed of a plurality of voxels each having an initial degree of freedom, the simulation space encompassing one or more interfaces of the component; defining an initial structure for the one or more interfaces in the simulation space; calculating, using a computer system with a finite floating-point precision, values for an electromagnetic field at each voxel using a finite-difference time domain solver to solve Maxwell's equations; and determining, for each voxel, whether to increase a respective numerical precision of respective values representing behavior of the electromagnetic field at the voxel above a threshold precision by the computer system and, in response, assigning one or more additional degrees of freedom to the voxel.

    Methods for designing hybrid neural networks having physical and digital components

    公开(公告)号:US11604957B1

    公开(公告)日:2023-03-14

    申请号:US16576946

    申请日:2019-09-20

    Abstract: Systems and methods for designing a hybrid neural network comprising at least one physical neural network component and at least one digital neural network component. A loss function is defined within a design space composed of a plurality of voxels, the design space encompassing one or more physical structures of the at least one physical neural network component and one or more architectural features of the digital neural network. Values are determined for at least one functional parameter for the one or more physical structures, and the at least one architectural parameter for the one or more architectural features, using a domain solver to solve Maxwell's equations so that a loss determined according to the loss function is within a threshold loss. Final structures are defined for the at least one physical neural network component and the digital neural network component based on the values.

    Multi-channel integrated photonic wavelength demultiplexer

    公开(公告)号:US11258527B2

    公开(公告)日:2022-02-22

    申请号:US17084032

    申请日:2020-10-29

    Abstract: A multi-channel photonic demultiplexer includes an input region to receive a multi-channel optical signal including four distinct wavelength channels, four output regions, each adapted to receive a corresponding one of the four distinct wavelength channels demultiplexed from the multi-channel optical signal, and a dispersive region optically disposed between the input region and the four output regions. The dispersive region includes a first material and a second material inhomogeneously interspersed to form a plurality of interfaces that each correspond to a change in refractive index of the dispersive region and collectively structure the dispersive region to optically separate each of the four distinct wavelength channels from the multi-channel optical signal and respectively guide each of the four distinct wavelength channels to the corresponding one of the four output regions.

    Physical device optimization with reduced computational latency via low-rank objectives

    公开(公告)号:US11238190B1

    公开(公告)日:2022-02-01

    申请号:US16392430

    申请日:2019-04-23

    Abstract: A method and system for optimizing structural parameters of a physical device is described. The method includes receiving an initial description of the physical device that describes structural parameters of the physical device within a simulated environment. The method further includes performing an operational simulation of the physical device in response to an excitation source, performing an adjoint simulation by backpropagating a placeholder metric through a simulated environment to determine a loss gradient, updating the loss gradient based, at least in part, on a loss metric determined from the operational simulation. Additionally, the method further comprises computing a structural gradient corresponding to an influence of changes in the structural parameters on the loss metric and generating a revised description of the physical device by updating the structural parameters based on the structural gradient to reduce the loss metric.

Patent Agency Ranking