公开(公告)号：US20220012058A1

公开(公告)日：2022-01-13

申请号：US17484780

申请日：2021-09-24

Applicant: Intel Corporation

Inventor： Niall Hanrahan ,  Martin Power ,  Kevin Brady ,  Martin-Thomas Grymel ,  David Bernard ,  Gary Baugh ,  Cormac Brick

IPC: G06F9/30 ,  G06F9/38 ,  G06F7/544

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed that increase data reuse for multiply and accumulate (MAC) operations. An example apparatus includes a MAC circuit to process a first context of a set of a first type of contexts stored in a first buffer and a first context of a set of a second type of contexts stored in a second buffer. The example apparatus also includes control logic circuitry to, in response to determining that there is an additional context of the second type to be processed in the set of the second type of contexts, maintain the first context of the first type in the first buffer. The control logic circuitry is also to, in response to determining that there is an additional context of the first type to be processed in the set of the first type of contexts maintain the first context of the second type in the second buffer and iterate a pointer of the second buffer from a first position to a next position in the second buffer.

公开(公告)号：US20240036763A1

公开(公告)日：2024-02-01

申请号：US18465560

申请日：2023-09-12

Applicant: Intel Corporation

Inventor： Niall Hanrahan ,  Martin Power ,  Kevin Brady ,  Martin-Thomas Grymel ,  David Bernard ,  Gary Baugh ,  Cormac Brick

IPC: G06F3/06 ,  G06F7/544

CPC classification number: G06F3/0656 ,  G06F7/5443 ,  G06F3/0625 ,  G06F3/0679 ,  G06F3/0613

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed that increase data reuse for multiply and accumulate (MAC) operations. An example apparatus includes a MAC circuit to process a first context of a set of a first type of contexts stored in a first buffer and a first context of a set of a second type of contexts stored in a second buffer. The example apparatus also includes control logic circuitry to, in response to determining that there is an additional context of the second type to be processed in the set of the second type of contexts, maintain the first context of the first type in the first buffer. The control logic circuitry is also to, in response to determining that there is an additional context of the first type to be processed in the set of the first type of contexts maintain the first context of the second type in the second buffer and iterate a pointer of the second buffer from a first position to a next position in the second buffer.

公开(公告)号：US20240403616A1

公开(公告)日：2024-12-05

申请号：US18500229

申请日：2023-11-02

Applicant: Intel Corporation

Inventor： Umer Iftikhar Cheema ,  Kevin Brady ,  Robert Simofi ,  Colm O Faolain ,  Deepak Abraham Mathaikutty ,  Arnab Raha ,  Dinakar Kondru ,  Gary Baugh ,  Darren Crews ,  Fergal Connor

IPC: G06N3/048

Abstract: An activation function in a neural network may be approximated by one or more linear functions. A linear function may correspond to a segment of the input range of the activation function, e.g., a linear segment. A programmable look-up table may store slopes and intercepts of linear functions. A post processing engine (PPE) array executing the activation function may determine that an input data element of the activation function falls into the linear segment and compute an output of the linear function using the input data element. The output of the linear function may be used as the approximated output of the activation function. Alternatively, the PPE array may determine that the input data element is in a saturation segment and use a fixed value associated with the saturation segment as the approximated output of the activation function.

公开(公告)号：US20220300418A1

公开(公告)日：2022-09-22

申请号：US17836720

申请日：2022-06-09

Applicant: Intel Corporation

Inventor： Gary Baugh

IPC: G06F12/02

Abstract: The present disclosure is discusses temporal access arbitration techniques for shared resources. Two separate address spaces may be defined for the shared resources and individual access agents. The temporal access arbitration techniques include temporally mapping addresses in an access agent address space to one or more addresses in the shared resource address space. The shared resources are accessed via linear addressing, where multiple addresses map to the same resources. Implementation constraints lead to a single resource being able to service several possible access agents per transaction cycle. In these ways, the temporal access arbitration techniques choreograph the access patterns of individual access agents so maximum resource bandwidth is achieved.

公开(公告)号：US11940907B2

公开(公告)日：2024-03-26

申请号：US17359217

申请日：2021-06-25

Applicant: Intel Corporation

Inventor： Martin-Thomas Grymel ,  David Bernard ,  Niall Hanrahan ,  Martin Power ,  Kevin Brady ,  Gary Baugh ,  Cormac Brick

IPC: G06F12/00 ,  G06F12/02 ,  G06N3/10

CPC classification number: G06F12/0207 ,  G06F12/0292 ,  G06N3/10

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed for sparse tensor storage for neural network accelerators. An example apparatus includes sparsity map generating circuitry to generate a sparsity map corresponding to a tensor, the sparsity map to indicate whether a data point of the tensor is zero, static storage controlling circuitry to divide the tensor into one or more storage elements, and a compressor to perform a first compression of the one or more storage elements to generate one or more compressed storage elements, the first compression to remove zero points of the one or more storage elements based on the sparsity map and perform a second compression of the one or more compressed storage elements, the second compression to store the one or more compressed storage elements contiguously in memory.

公开(公告)号：US11789646B2

公开(公告)日：2023-10-17

申请号：US17484780

申请日：2021-09-24

Applicant: Intel Corporation

Inventor： Niall Hanrahan ,  Martin Power ,  Kevin Brady ,  Martin-Thomas Grymel ,  David Bernard ,  Gary Baugh ,  Cormac Brick

IPC: G06F3/06 ,  G06F7/544

CPC classification number: G06F3/0656 ,  G06F3/0613 ,  G06F3/0625 ,  G06F3/0679 ,  G06F7/5443

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed that increase data reuse for multiply and accumulate (MAC) operations. An example apparatus includes a MAC circuit to process a first context of a set of a first type of contexts stored in a first buffer and a first context of a set of a second type of contexts stored in a second buffer. The example apparatus also includes control logic circuitry to, in response to determining that there is an additional context of the second type to be processed in the set of the second type of contexts, maintain the first context of the first type in the first buffer. The control logic circuitry is also to, in response to determining that there is an additional context of the first type to be processed in the set of the first type of contexts maintain the first context of the second type in the second buffer and iterate a pointer of the second buffer from a first position to a next position in the second buffer.

公开(公告)号：US12169643B2

公开(公告)日：2024-12-17

申请号：US18465560

申请日：2023-09-12

Applicant: Intel Corporation

Inventor： Niall Hanrahan ,  Martin Power ,  Kevin Brady ,  Martin-Thomas Grymel ,  David Bernard ,  Gary Baugh ,  Cormac Brick

IPC: G06F3/06 ,  G06F7/544

Abstract: Methods, apparatus, systems, and articles of manufacture are disclosed that increase data reuse for multiply and accumulate (MAC) operations. An example apparatus includes a MAC circuit to process a first context of a set of a first type of contexts stored in a first buffer and a first context of a set of a second type of contexts stored in a second buffer. The example apparatus also includes control logic circuitry to, in response to determining that there is an additional context of the second type to be processed in the set of the second type of contexts, maintain the first context of the first type in the first buffer. The control logic circuitry is also to, in response to determining that there is an additional context of the first type to be processed in the set of the first type of contexts maintain the first context of the second type in the second buffer and iterate a pointer of the second buffer from a first position to a next position in the second buffer.

公开(公告)号：US20240134786A1

公开(公告)日：2024-04-25

申请号：US18539955

申请日：2023-12-14

Applicant: Intel Corporation

Inventor： Martin-Thomas Grymel ,  David Bernard ,  Niall Hanrahan ,  Martin Power ,  Kevin Brady ,  Gary Baugh ,  Cormac Brick

IPC: G06F12/02 ,  G06N3/10

CPC classification number: G06F12/0207 ,  G06F12/0292 ,  G06N3/10

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed for sparse tensor storage for neural network accelerators. An example apparatus includes sparsity map generating circuitry to generate a sparsity map corresponding to a tensor, the sparsity map to indicate whether a data point of the tensor is zero, static storage controlling circuitry to divide the tensor into one or more storage elements, and a compressor to perform a first compression of the one or more storage elements to generate one or more compressed storage elements, the first compression to remove zero points of the one or more storage elements based on the sparsity map and perform a second compression of the one or more compressed storage elements, the second compression to store the one or more compressed storage elements contiguously in memory.

公开(公告)号：US20210406164A1

公开(公告)日：2021-12-30

申请号：US17359217

申请日：2021-06-25

Applicant: Intel Corporation

Inventor： Martin-Thomas Grymel ,  David Bernard ,  Niall Hanrahan ,  Martin Power ,  Kevin Brady ,  Gary Baugh ,  Cormac Brick

IPC: G06F12/02 ,  G06N3/10

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed for sparse tensor storage for neural network accelerators. An example apparatus includes sparsity map generating circuitry to generate a sparsity map corresponding to a tensor, the sparsity map to indicate whether a data point of the tensor is zero, static storage controlling circuitry to divide the tensor into one or more storage elements, and a compressor to perform a first compression of the one or more storage elements to generate one or more compressed storage elements, the first compression to remove zero points of the one or more storage elements based on the sparsity map and perform a second compression of the one or more compressed storage elements, the second compression to store the one or more compressed storage elements contiguously in memory.

公开(公告)号：US20210319317A1

公开(公告)日：2021-10-14

申请号：US17357924

申请日：2021-06-24

Applicant: Intel Corporation

Inventor： Martin Power ,  Kevin Brady ,  Niall Hanrahan ,  Martin-Thomas Grymel ,  David Bernard ,  Gary Baugh

IPC: G06N3/08 ,  G06N3/063

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to perform machine-learning model operations on sparse accelerators. An example apparatus includes first circuitry, second circuitry to generate sparsity data based on an acceleration operation, and third circuitry to instruct one or more data buffers to provide at least one of activation data or weight data based on the sparsity data to the first circuitry, the first circuitry to execute the acceleration operation based on the at least one of the activation data or the weight data.