公开(公告)号：US11526736B2

公开(公告)日：2022-12-13

申请号：US16541878

申请日：2019-08-15

Applicant: Intel Corporation

Inventor： Estelle Aflalo ,  Amit Bleiweiss ,  Mattias Marder ,  Eliran Zimmerman

IPC: G06E1/00 ,  G06G7/00 ,  G06N3/063 ,  G06K9/62 ,  G06N3/08 ,  G06F9/50

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to map workloads. An example apparatus includes a constraint definer to define performance characteristic targets of the neural network, an action determiner to apply a first resource configuration to candidate resources corresponding to the neural network, a reward determiner to calculate a results metric based on (a) resource performance metrics and (b) the performance characteristic targets, and a layer map generator to generate a resource mapping file, the mapping file including respective resource assignments for respective corresponding layers of the neural network, the resource assignments selected based on the results metric.

公开(公告)号：US11816561B2

公开(公告)日：2023-11-14

申请号：US17977774

申请日：2022-10-31

Applicant: Intel Corporation

Inventor： Estelle Aflalo ,  Amit Bleiweiss ,  Mattias Marder ,  Eliran Zimmerman

IPC: G06N3/08 ,  G06N3/063 ,  G06F9/50 ,  G06F18/21

CPC classification number: G06N3/063 ,  G06F9/5011 ,  G06F9/5044 ,  G06F18/217 ,  G06N3/08

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to map workloads. An example apparatus includes a constraint definer to define performance characteristic targets of the neural network, an action determiner to apply a first resource configuration to candidate resources corresponding to the neural network, a reward determiner to calculate a results metric based on (a) resource performance metrics and (b) the performance characteristic targets, and a layer map generator to generate a resource mapping file, the mapping file including respective resource assignments for respective corresponding layers of the neural network, the resource assignments selected based on the results metric.

公开(公告)号：US20250124271A1

公开(公告)日：2025-04-17

申请号：US18999998

申请日：2024-12-23

Applicant: Intel Corporation

Inventor： Estelle Aflalo ,  Amit Bleiweiss ,  Mattias Marder ,  Eliran Zimmerman

IPC: G06N3/063 ,  G06F9/50 ,  G06F18/21 ,  G06N3/08

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to map workloads. An example apparatus includes a constraint definer to define performance characteristic targets of the neural network, an action determiner to apply a first resource configuration to candidate resources corresponding to the neural network, a reward determiner to calculate a results metric based on (a) resource performance metrics and (b) the performance characteristic targets, and a layer map generator to generate a resource mapping file, the mapping file including respective resource assignments for respective corresponding layers of the neural network, the resource assignments selected based on the results metric.

公开(公告)号：US12093836B2

公开(公告)日：2024-09-17

申请号：US17129521

申请日：2020-12-21

Applicant: Intel Corporation

Inventor： Mattias Marder ,  Estelle Aflalo ,  Avrech Ben-David ,  Shauharda Khadka ,  Somdeb Majumdar ,  Santiago Miret ,  Hanlin Tang

IPC: G06N3/10 ,  G06N3/045 ,  G06N3/086

CPC classification number: G06N3/10 ,  G06N3/045 ,  G06N3/086

Abstract: Automatic multi-objective hardware optimization for processing a deep learning network is disclosed. An example of a storage medium includes instructions for obtaining client preferences for a plurality of performance indicators for processing of a deep learning workload; generating a workload representation for the deep learning workload; providing the workload representation to machine learning processing to generate a workload executable, the workload executable including hardware mapping based on the client preferences; and applying the workload executable in processing of the deep learning workload.

公开(公告)号：US20240119271A1

公开(公告)日：2024-04-11

申请号：US18491246

申请日：2023-10-20

Applicant: Intel Corporation

Inventor： Estelle Aflalo ,  Amit Bleiweiss ,  Mattias Marder ,  Eliran Zimmerman

IPC: G06N3/063 ,  G06F9/50 ,  G06F18/21 ,  G06N3/08

CPC classification number: G06N3/063 ,  G06F9/5011 ,  G06F9/5044 ,  G06F18/217 ,  G06N3/08

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to map workloads. An example apparatus includes a constraint definer to define performance characteristic targets of the neural network, an action determiner to apply a first resource configuration to candidate resources corresponding to the neural network, a reward determiner to calculate a results metric based on (a) resource performance metrics and (b) the performance characteristic targets, and a layer map generator to generate a resource mapping file, the mapping file including respective resource assignments for respective corresponding layers of the neural network, the resource assignments selected based on the results metric.

公开(公告)号：US20230111365A1

公开(公告)日：2023-04-13

申请号：US17977774

申请日：2022-10-31

Applicant: Intel Corporation

Inventor： Estelle Aflalo ,  Amit Bleiweiss ,  Mattias Marder ,  Eliran Zimmerman

IPC: G06N3/063 ,  G06N3/08 ,  G06F9/50 ,  G06F18/21

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to map workloads. An example apparatus includes a constraint definer to define performance characteristic targets of the neural network, an action determiner to apply a first resource configuration to candidate resources corresponding to the neural network, a reward determiner to calculate a results metric based on (a) resource performance metrics and (b) the performance characteristic targets, and a layer map generator to generate a resource mapping file, the mapping file including respective resource assignments for respective corresponding layers of the neural network, the resource assignments selected based on the results metric.

公开(公告)号：US12260323B2

公开(公告)日：2025-03-25

申请号：US18491246

申请日：2023-10-20

Applicant: Intel Corporation

Inventor： Estelle Aflalo ,  Amit Bleiweiss ,  Mattias Marder ,  Eliran Zimmerman

IPC: G06N3/08 ,  G06F9/50 ,  G06F18/21 ,  G06N3/063

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed to map workloads. An example apparatus includes a constraint definer to define performance characteristic targets of the neural network, an action determiner to apply a first resource configuration to candidate resources corresponding to the neural network, a reward determiner to calculate a results metric based on (a) resource performance metrics and (b) the performance characteristic targets, and a layer map generator to generate a resource mapping file, the mapping file including respective resource assignments for respective corresponding layers of the neural network, the resource assignments selected based on the results metric.

公开(公告)号：US20210150371A1

公开(公告)日：2021-05-20

申请号：US17129521

申请日：2020-12-21

Applicant: Intel Corporation

Inventor： Mattias Marder ,  Estelle Aflalo ,  Avrech Ben-David ,  Shauharda Khadka ,  Somdeb Majumdar ,  Santiago Miret ,  Hanlin Tang

IPC: G06N3/10 ,  G06N3/08 ,  G06N3/04

Abstract: Automatic multi-objective hardware optimization for processing a deep learning network is disclosed. An example of a storage medium includes instructions for obtaining client preferences for a plurality of performance indicators for processing of a deep learning workload; generating a workload representation for the deep learning workload; providing the workload representation to machine learning processing to generate a workload executable, the workload executable including hardware mapping based on the client preferences; and applying the workload executable in processing of the deep learning workload.