Method and apparatus with neural network processing

    公开(公告)号:US11410410B2

    公开(公告)日:2022-08-09

    申请号:US16906326

    申请日:2020-06-19

    Inventor: Sangwon Ha

    Abstract: A method of processing a neural network, includes generating an integral map for each channel in a first layer of the neural network based on calculating of area sums of pixel values in first output feature maps of channels in the first layer, generating an accumulated integral map by performing an accumulation operation on the integral maps generated for the respective channels, obtaining pre-output feature maps of a second layer, subsequent to the first layer, by performing a convolution operation between input feature maps of the second layer and weight kernels, and removing offsets in the weight kernels to obtain second output feature maps of the second layer by subtracting accumulated values of the accumulated integral map from pixel values of the pre-output feature maps.

    Method and apparatus with neural network data quantizing

    公开(公告)号:US12106219B2

    公开(公告)日:2024-10-01

    申请号:US15931362

    申请日:2020-05-13

    CPC classification number: G06N3/084 G06N3/04 G06N3/0495

    Abstract: A neural network data quantizing method includes: obtaining local quantization data by firstly quantizing, based on a local maximum value for each output channel of a current layer of a neural network, global recovery data obtained by recovering output data of an operation of the current layer based on a global maximum value corresponding to a previous layer of the neural network; storing the local quantization data in a memory to perform an operation of a next layer of the neural network; obtaining global quantization data by secondarily quantizing, based on a global maximum value corresponding to the current layer, local recovery data obtained by recovering the local quantization data based on the local maximum value for each output channel of the current layer; and providing the global quantization data as input data for the operation of the next layer.

    Method and apparatus with neural network processing

    公开(公告)号:US11823029B2

    公开(公告)日:2023-11-21

    申请号:US17846202

    申请日:2022-06-22

    Inventor: Sangwon Ha

    CPC classification number: G06N3/048 G06F18/21 G06N3/04 G06N3/08 G06N20/10

    Abstract: A method of processing a neural network, includes generating an integral map for each channel in a first layer of the neural network based on calculating of area sums of pixel values in first output feature maps of channels in the first layer, generating an accumulated integral map by performing an accumulation operation on the integral maps generated for the respective channels, obtaining pre-output feature maps of a second layer, subsequent to the first layer, by performing a convolution operation between input feature maps of the second layer and weight kernels, and removing offsets in the weight kernels to obtain second output feature maps of the second layer by subtracting accumulated values of the accumulated integral map from pixel values of the pre-output feature maps.

    Method and apparatus with neural network parameter quantization

    公开(公告)号:US11816557B2

    公开(公告)日:2023-11-14

    申请号:US17950342

    申请日:2022-09-22

    Abstract: A processor-implemented neural network method includes: determining a respective probability density function (PDF) of normalizing a statistical distribution of parameter values, for each channel of each of a plurality of feature maps of a pre-trained neural network; determining, for each channel, a corresponding first quantization range for performing quantization of corresponding parameter values, based on a quantization error and a quantization noise of the respective determined PDF; determining, for each channel, a corresponding second quantization range, based on a signal-to-quantization noise ratio (SQNR) of the respective determined PDF; correcting, for each channel, the corresponding first quantization range based on the corresponding second quantization range; and generating a quantized neural network, based on the corrected first quantization range corresponding for each channel.

    Circuit board including insulating layer having a plurality of dielectrics with different dielectric loss, and electronic device including the circuit board

    公开(公告)号:US11605891B2

    公开(公告)日:2023-03-14

    申请号:US16460937

    申请日:2019-07-02

    Abstract: An electronic device includes a communication circuit electrically connected with a circuit board. The circuit board includes a first portion comprising a first layered structure in which a wiring layer and a first insulating layer are alternately positioned, and a second portion comprising a second layered structure in which the wiring layer and the first insulating layer are alternately positioned and a second insulating layer. At least one antenna patch is positioned on or within the second insulating layer. A conductive line penetrates the second layered structure and the second insulating layer and electrically connects the at least one antenna patch and the communication circuit. The first insulating layer has a first loss tangent value, and the second insulating layer has a second loss tangent value smaller than the first loss tangent value.

    Electronic device including interposer

    公开(公告)号:US11317512B2

    公开(公告)日:2022-04-26

    申请号:US16883604

    申请日:2020-05-26

    Abstract: An electronic device is provided that includes a first circuit board including a first electronic component and a second electronic component disposed on a side of the first circuit board, a second circuit board spaced apart from the first circuit board and having a side facing the side of the first circuit board on which the first electronic component and the second electronic component are disposed, a first interposer disposed between the first circuit board and the second circuit board to form an inner space between the first circuit board and the second circuit board, and a second interposer disposed between the first circuit board and the second circuit board to divide the inner space into a first region and a second region, wherein the first electronic component on the first circuit board is disposed corresponding to the first region, and the second electronic component on the first circuit board is disposed corresponding to the second region, and wherein the first interposer and the second interposer electrically connect the first circuit board to the second circuit board.

    CABLE DEVICE
    18.
    发明申请

    公开(公告)号:US20210111520A1

    公开(公告)日:2021-04-15

    申请号:US17008052

    申请日:2020-08-31

    Abstract: Provided is a cable device having an improved electromagnetic interference (EMI) shielding performance. The cable device includes: a cable including an optical fiber; a connector including a printed circuit board connected to the cable and including a ground electrode, and a conductive case; a connecting member provided around a connection between the cable and the connector; and a metal shell surrounding the cable inside of the connecting member, the metal shell being configured to shield electromagnetic interference of the cable and the connector.

    METHOD AND APPARATUS WITH NEURAL NETWORK DATA QUANTIZING

    公开(公告)号:US20210110270A1

    公开(公告)日:2021-04-15

    申请号:US15931362

    申请日:2020-05-13

    Abstract: A neural network data quantizing method includes: obtaining local quantization data by firstly quantizing, based on a local maximum value for each output channel of a current layer of a neural network, global recovery data obtained by recovering output data of an operation of the current layer based on a global maximum value corresponding to a previous layer of the neural network; storing the local quantization data in a memory to perform an operation of a next layer of the neural network; obtaining global quantization data by secondarily quantizing, based on a global maximum value corresponding to the current layer, local recovery data obtained by recovering the local quantization data based on the local maximum value for each output channel of the current layer; and providing the global quantization data as input data for the operation of the next layer.

Patent Agency Ranking