公开(公告)号：US11995472B2

公开(公告)日：2024-05-28

申请号：US17378841

申请日：2021-07-19

Applicant: Texas Instruments Incorporated

Inventor： Mihir Narendra Mody ,  Kedar Satish Chitnis ,  Kumar Desappan ,  David Smith ,  Pramod Kumar Swami ,  Shyam Jagannathan

IPC: G06F9/50 ,  G06F9/455 ,  G06F12/00 ,  G06F12/02 ,  G06N3/02 ,  G06N3/10 ,  G06N20/00

CPC classification number: G06F9/5016 ,  G06F9/5077 ,  G06F12/00 ,  G06F12/0223 ,  G06F2009/45583 ,  G06F9/50 ,  G06F9/5022 ,  G06N3/02 ,  G06N3/10 ,  G06N20/00

Abstract: Techniques for executing machine learning (ML) models including receiving an indication to run an ML model on a processing core; receiving a static memory allocation for running the ML model on the processing core; determining that a layer of the ML model uses more memory than the static memory allocated; transmitting, to a shared memory, a memory request for blocks of the shared memory; receiving an allocation of the requested blocks; running the layer of the ML model using the static memory and the range of memory addresses; and outputting results of running the layer of the ML model.

公开(公告)号：US20220295079A1

公开(公告)日：2022-09-15

申请号：US17827988

申请日：2022-05-30

Applicant: TEXAS INSTRUMENTS INCORPORATED

Inventor： Yashwant Dutt ,  Kumar Desappan ,  Piyali Goswami

IPC: H04N19/167 ,  H04N19/176 ,  H04N19/103 ,  H04N19/124 ,  H04N19/157

Abstract: Several methods and systems for masking multimedia data are disclosed. In an embodiment, a method for masking includes performing a prediction for at least one multimedia data block based on a prediction mode of a plurality of prediction modes. The at least one multimedia data block is associated with a region of interest (ROI). A residual multimedia data associated with the at least one multimedia data block is generated based on the prediction. A quantization of the residual multimedia data is performed based on a quantization parameter (QP) value. The QP value is variable such that varying the QP value controls a degree of masking of the ROI.

公开(公告)号：US11048997B2

公开(公告)日：2021-06-29

申请号：US15800294

申请日：2017-11-01

Applicant: Texas Instruments Incorporated

Inventor： Manu Mathew ,  Kumar Desappan ,  Pramod Kumar Swami

IPC: G06N3/04 ,  G06N3/08 ,  G06F7/544 ,  G06F17/15 ,  G06N3/10

Abstract: A method for convolution in a convolutional neural network (CNN) is provided that includes accessing a coefficient value of a filter corresponding to an input feature map of a convolution layer of the CNN, and performing a block multiply accumulation operation on a block of data elements of the input feature map, the block of data elements corresponding to the coefficient value, wherein, for each data element of the block of data elements, a value of the data element is multiplied by the coefficient value and a result of the multiply is added to a corresponding data element in a corresponding output block of data elements comprised in an output feature map.

公开(公告)号：US20190116364A1

公开(公告)日：2019-04-18

申请号：US16213527

申请日：2018-12-07

Applicant: TEXAS INSTRUMENTS INCORPORATED

Inventor： Yashwant Dutt ,  Kumar Desappan ,  Piyali Goswami

IPC: H04N19/167 ,  H04N19/157 ,  H04N19/176 ,  H04N19/103 ,  H04N19/124

Abstract: Several methods and systems for masking multimedia data are disclosed. In an embodiment, a method for masking includes performing a prediction for at least one multimedia data block based on a prediction mode of a plurality of prediction modes. The at least one multimedia data block is associated with a region of interest (ROI). A residual multimedia data associated with the at least one multimedia data block is generated based on the prediction. A quantization of the residual multimedia data is performed based on a quantization parameter (QP) value. The QP value is variable such that varying the QP value controls a degree of masking of the ROI.

公开(公告)号：US20180181857A1

公开(公告)日：2018-06-28

申请号：US15800294

申请日：2017-11-01

Applicant: Texas Instruments Incorporated

Inventor： Manu Mathew ,  Kumar Desappan ,  Pramod Kumar Swami

IPC: G06N3/04 ,  G06N3/08

CPC classification number: G06N3/04 ,  G06N3/0454 ,  G06N3/08 ,  G06N3/082 ,  G06N3/084 ,  G06N3/10

Abstract: A method for convolution in a convolutional neural network (CNN) is provided that includes accessing a coefficient value of a filter corresponding to an input feature map of a convolution layer of the CNN, and performing a block multiply accumulation operation on a block of data elements of the input feature map, the block of data elements corresponding to the coefficient value, wherein, for each data element of the block of data elements, a value of the data element is multiplied by the coefficient value and a result of the multiply is added to a corresponding data element in a corresponding output block of data elements comprised in an output feature map.

公开(公告)号：US12099930B2

公开(公告)日：2024-09-24

申请号：US17117271

申请日：2020-12-10

Applicant: Texas Instruments Incorporated

Inventor： Manu Mathew ,  Kumar Desappan ,  Soyeb Noormohammed Nagori ,  Debapriya Maji ,  Pramod Kumar Swami

IPC: G06N3/048 ,  G06N3/04 ,  G06N3/084

CPC classification number: G06N3/084 ,  G06N3/04 ,  G06N3/048

Abstract: In described examples of a method for quantizing data for a convolutional neural network (CNN) is provided. A set of data is received and quantized the using a power-of-2 parametric activation (PACT2) function. The PACT2 function arranges the set of data as a histogram and discards a portion of the data corresponding to a tail of the histogram to form a remaining set of data. A clipping value is determined by expanding the remaining set of data to a nearest power of two value. The set of data is then quantized using the clipping value. With PACT2, a model can be quantized either using post training quantization or using quantization aware training. PACT2 helps a quantized model to achieve close accuracy compared to the corresponding floating-point model.

公开(公告)号：US20240036816A1

公开(公告)日：2024-02-01

申请号：US18193396

申请日：2023-03-30

Applicant: TEXAS INSTRUMENTS INCORPORATED

Inventor： Kumar Desappan ,  Anshu Jain ,  Manu Mathew

IPC: G06F5/01 ,  G06F7/485

CPC classification number: G06F5/012 ,  G06F7/485

Abstract: Disclosed herein are systems and methods for determining the scaling factors for a neural network that satisfy the activation functions employed by the nodes of the network. A processor identifies a saturation point of an activation function. Next, the processor determines a scaling factor for an output feature map based on the saturation point of the activation function. Then, the processor determines a scaling factor for an accumulator based on the scaling for the output feature map and further based on a shift value related to a quantization. Finally, the processor determines a scaling factor for a weight map based on the scaling factor for the accumulator.

公开(公告)号：US11580719B2

公开(公告)日：2023-02-14

申请号：US17128365

申请日：2020-12-21

Applicant: Texas Instruments Incorporated

Inventor： Kumar Desappan ,  Manu Mathew ,  Pramod Kumar Swami ,  Praveen Eppa

IPC: G06V10/28 ,  G06K9/62 ,  G06N3/04 ,  G06N3/08 ,  G06N3/063 ,  G06V10/44

Abstract: A method for dynamically quantizing feature maps of a received image. The method includes convolving an image based on a predicted maximum value, a predicted minimum value, trained kernel weights and the image data. The input data is quantized based on the predicted minimum value and predicted maximum value. The output of the convolution is computed into an accumulator and re-quantized. The re-quantized value is output to an external memory. The predicted min value and the predicted max value are computed based on the previous max values and min values with a weighted average or a pre-determined formula. Initial min value and max value are computed based on known quantization methods and utilized for initializing the predicted min value and predicted max value in the quantization process.

公开(公告)号：US20220327355A1

公开(公告)日：2022-10-13

申请号：US17809677

申请日：2022-06-29

Applicant: Texas Instruments Incorporated

Inventor： Manu Mathew ,  Kumar Desappan ,  Pramod Kumar Swami

IPC: G06N3/04 ,  G06N3/08 ,  G06F17/15 ,  G06F7/544

Abstract: A method for generating a sparsified convolutional neural network (CNN) is provided that includes training the CNN to generate coefficient values of filters of convolution layers, and performing sparsified fine tuning on the convolution layers to generate the sparsified CNN, wherein the sparsified fine tuning causes selected nonzero coefficient values of the filters to be set to zero.

公开(公告)号：US20210279550A1

公开(公告)日：2021-09-09

申请号：US17327988

申请日：2021-05-24

Applicant: TEXAS INSTRUMENTS INCORPORATED

Inventor： Manu Mathew ,  Kumar Desappan ,  Pramod Kumar Swami

IPC: G06N3/04 ,  G06N3/08 ,  G06F17/15 ,  G06F7/544

Abstract: A method for convolution in a convolutional neural network (CNN) is provided that includes accessing a coefficient value of a filter corresponding to an input feature map of a convolution layer of the CNN, and performing a block multiply accumulation operation on a block of data elements of the input feature map, the block of data elements corresponding to the coefficient value, wherein, for each data element of the block of data elements, a value of the data element is multiplied by the coefficient value and a result of the multiply is added to a corresponding data element in a corresponding output block of data elements comprised in an output feature map.