US 12,450,419 B1
Automatic design parameter optimization for electronic circuit designs with operating environment coverage
Xiaopeng Duan, Austin, TX (US)
Assigned to Synopsys, Inc., Sunnyvale, CA (US)
Filed by Synopsys, Inc., Mountain View, CA (US)
Filed on Jun. 13, 2022, as Appl. No. 17/838,671.
Claims priority of provisional application 63/210,400, filed on Jun. 14, 2021.
Int. Cl. G06F 30/398 (2020.01); G06F 30/27 (2020.01); G06F 119/02 (2020.01)
CPC G06F 30/398 (2020.01) [G06F 30/27 (2020.01); G06F 2119/02 (2020.01)] 20 Claims
OG exemplary drawing
 
1. A method for optimizing parameters of a circuit design, the method comprising:
receiving a circuit design;
accessing a model configured to receive design variables of the circuit design and predict a quality of the circuit design;
for each of a plurality of coverages, performing:
generating a set of samples for the coverage, each sample representing a set of values of design parameters;
for each sample from the set of samples, predicting the quality of the sample using the model;
selecting a subset of samples determined to have a predicted quality that exceeds a target;
performing simulations of the selected subset of samples; and
performing incremental training of the model based on results of the simulations of the selected subset of samples; and
selecting one or more samples based on results of the simulations.