On Approaching Multivariate IC Pre-silicon Verification Using ML-based Adaptive Algorithms

This paper introduces several solutions for multivariate extension of a previously designed single response adaptive pre-silicon integrated circuit verification approach employing machine learning algorithms. These techniques aim to achieve the most accurate identification of worst-case circuit behavior through simultaneously modeling multiple electrical parameters (EP). The effectiveness of the proposed methods was validated through extensive testing on a large and diverse set of synthetic test functions that intend to replicate the behavior of real circuits. The algorithms consistency and accuracy are also validated on a real Low Dropout Voltage Regulator (LDO) circuit.