Multi-dimension variable predictive modeling for analysis acceleration
In one example, a method for evaluating a system includes obtaining a model of the system that defines a boundary between at least one failure region and a non-failure region for a performance indicator with respect to at least one variable of the system. In one embodiment, obtaining the model involves constructing a new model; however, in other embodiments, obtaining the model involves accepting or retrieving a pre-constructed model is input. The method further includes obtaining importance samples for the at least one variable that are biased to the at least one failure region, and calculating indicator values for the performance indicator by applying the importance samples to the model.
1. A method, comprising:
obtaining a model of an integrated circuit, wherein the model defines a boundary between at least one failure region and a non-failure region for a performance indicator with respect to at least one variable of the integrated circuit, wherein the model comprises a decision tree that models a relationship between the at least one variable and the at least one failure region;
obtaining importance samples for the at least one variable, wherein the importance samples are biased to the at least one failure region;
calculating indicator values for the performance indicator of the integrated circuit by applying the importance samples to the model integrated circuit;
generating a predicted integrated circuit yield for a design of the integrated circuit, based on the indicator values;
adjusting the at least one variable based on the predicted integrated circuit yield; and
fabricating the integrated circuit according to the design, subsequent to the adjusting.
2. The method of claim 1 , wherein a target of the decision tree is the at least one failure region, and an input of the decision tree is the at least one variable.
3. The method of claim 1 , wherein the relationship is modeled as a sequence of decisions being made based on values of the at least one variable.
4. The method of claim 1 , wherein the obtaining the model comprises constructing the model.
5. The method of claim 1 , wherein the obtaining the model comprises accepting or retrieving the model, wherein the model is pre-constructed.
6. A non-transitory computer-readable storage medium storing instructions which, when executed by the processor, cause the processor to perform operations, the operations comprising:
obtaining a model of an integrated circuit, wherein the model defines a boundary between at least one failure region and a non-failure region for a performance indicator with respect to at least one variable of the integrated circuit, wherein the model comprises a decision tree that models a relationship between the at least one variable and the at least one failure region;
obtaining importance samples for the at least one variable, wherein the importance samples are biased to the at least one failure region;
calculating indicator values for the performance indicator of the integrated circuit by applying the importance samples to the model integrated circuit;
generating a predicted integrated circuit yield for a design of the integrated circuit, based on the indicator values;
adjusting the at least one variable based on the predicted integrated circuit yield; and
fabricating the integrated circuit according to the design, subsequent to the adjusting.
7. The non-transitory computer-readable storage medium of claim 6 , wherein the relationship is modeled as a sequence of decisions being made based on values of the at least one variable.
8. The non-transitory computer-readable storage medium of claim 6 , wherein the obtaining the model comprises constructing the model.
9. The non-transitory computer-readable storage medium of claim 6 , wherein the obtaining the model comprises accepting or retrieving the model, wherein the model is pre-constructed.
10. The non-transitory computer-readable storage medium of claim 6 , wherein a target of the decision tree is the at least one failure region, and an input of the decision tree is the at least one variable.
11. A method, comprising:
obtaining a model of an integrated circuit, wherein the model defines a boundary between at least one failure region and a non-failure region for a performance indicator with respect to at least one variable of the integrated circuit, wherein the model comprises a multidimensional lookup table;
obtaining importance samples for the at least one variable, wherein the importance samples are biased to the at least one failure region;
calculating indicator values for the performance indicator of the integrated circuit by applying the importance samples to the model integrated circuit;
generating a predicted integrated circuit yield for a design of the integrated circuit, based on the indicator values;
adjusting the at least one variable based on the predicted integrated circuit yield; and
fabricating the integrated circuit according to the design, subsequent to the adjusting.
12. The method of claim 11 , wherein the obtaining the model comprises constructing the model.
13. The method of claim 11 , wherein the obtaining the model comprises accepting or retrieving the model, wherein the model is pre-constructed.
14. The method of claim 11 , wherein the obtaining the model comprises constructing the model.
15. The method of claim 11 , wherein the obtaining the model comprises accepting or retrieving the model, wherein the model is pre-constructed.