Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

2results about How to "Satisfy constraints" patented technology

A method for OPC processing of metal layer pattern (MRC) restricted

PendingCN122652882Afull coverageprevent leakage
The application relates to the technical field of semiconductor integrated circuit manufacturing, and particularly discloses an OPC processing method for solving the MRC limitation of metal layer patterns, which comprises the following steps: S1, screening out metal layer line ends with hole layers as to-be-optimized line ends from a metal layer layout; S2, identifying 'L' type line ends from the to-be-optimized line ends, and the patterns opposite to the 'L' type line ends are 'I' type line ends; S3, screening out short side 'L' type line ends from the 'L' type line ends, and the length of the short side of the short side 'L' type line end is less than 0.5 times of a design standard of a hole to a target; and S4, identifying adjacent sides of the short side, and generating line end marks according to the adjacent sides. The OPC processing method for solving the MRC limitation of metal layer patterns can ensure that the 'I' type line ends obtain sufficient compensation by preferentially controlling and setting edge placement errors of the contours to the connected adjacent sides, and solves the problem of contour retreat of the 'I' type line ends.
Owner:CHONGQING XINLIAN MICROELECTRONICS CO LTD

Neural architecture and hardware accelerator search

ActiveCN116324807BSupport network performancesatisfy constraintsPhysical realisationComputer hardwareAlgorithm
Methods, systems, and apparatus for jointly determining neural network architecture and hardware accelerator architecture, including computer programs encoded on a computer storage medium. In one aspect, a method includes: generating a batch of one or more output sequences using a controller policy, each output sequence in the batch defining a corresponding architecture of a sub-neural network and a corresponding architecture of a hardware accelerator; for each output sequence in the batch: training a corresponding instance of the sub-neural network having the architecture defined by the output sequence; evaluating the network performance of the trained instance of the sub-neural network; and evaluating the accelerator performance of a corresponding instance of the hardware accelerator having the architecture defined by the output sequence to determine an accelerator performance metric for the instance of the hardware accelerator; and adjusting the controller policy using the network performance metric and the accelerator performance metric.
Owner:GOOGLE LLC