公开(公告)号：US20230185987A1

公开(公告)日：2023-06-15

申请号：US17551514

申请日：2021-12-15

Applicant: X Development LLC

Inventor： Ardavan Oskooi ,  Martin Schubert ,  Cyril Pierre Bortolato

IPC: G06F30/20

CPC classification number: G06F30/20

Abstract: In some embodiments, a non-transitory computer-readable medium is provided. The computer-readable medium has logic stored thereon that, in response to execution by one or more processors of a computing system, cause the computing system to perform actions for deriving a fabrication model for a fabrication system using an inverse design process. The actions include determining a test design for a test physical device, measuring performance of an instance of the test physical device fabricated by the fabrication system using the test design to determine an as-fabricated performance metric, optimizing the test design using a first loss function based on differences in a simulated performance metric of the test design and the as-fabricated performance metric to determine an as-fabricated design, and optimizing a fabrication model using a second loss function based on differences between the test design and the as-fabricated design.

公开(公告)号：US11893484B2

公开(公告)日：2024-02-06

申请号：US17111364

申请日：2020-12-03

Applicant: X Development LLC

Inventor： Ardavan Oskooi ,  Christopher Hogan ,  Alec M. Hammond ,  Steven G. Johnson

IPC: G06F30/20 ,  G06N3/08 ,  G06N3/045

CPC classification number: G06N3/08 ,  G06F30/20 ,  G06N3/045

Abstract: In some embodiments, a method for optimal parallel execution of a simulation of a design is provided. A computing device extracts one or more features from the design. The computing device provides at least the one or more features as inputs to one or more machine learning models to determine one or more predictions of execution times. The computing device determines an optimum execution architecture based on the one or more predictions of execution times. The computing device distributes portions of the design for simulation based on the optimum execution architecture. In some embodiments, one or more machine learning models are trained to generate outputs for predicting an optimal parallel execution architecture for simulation of a design.

公开(公告)号：US20210174206A1

公开(公告)日：2021-06-10

申请号：US17111364

申请日：2020-12-03

Applicant: X Development LLC

Inventor： Ardavan Oskooi ,  Christopher Hogan ,  Alec M. Hammond ,  Steven G. Johnson

IPC: G06N3/08 ,  G06F30/20 ,  G06N3/04

Abstract: In some embodiments, a method for optimal parallel execution of a simulation of a design is provided. A computing device extracts one or more features from the design. The computing device provides at least the one or more features as inputs to one or more machine learning models to determine one or more predictions of execution times. The computing device determines an optimum execution architecture based on the one or more predictions of execution times. The computing device distributes portions of the design for simulation based on the optimum execution architecture. In some embodiments, one or more machine learning models are trained to generate outputs for predicting an optimal parallel execution architecture for simulation of a design.