Causal AI For AMS Circuit Design: Interpretable Parameter Effects Analysis

TL;DR

This paper introduces a causal-inference framework for AMS circuit design that improves interpretability and accuracy over traditional neural networks by quantifying parameter impacts and providing human-understandable insights.

Contribution

The paper presents a novel causal AI approach that discovers a DAG from simulation data and estimates parameter effects, enhancing interpretability and accuracy in AMS circuit modeling.

Findings

Causal model achieves less than 25% average absolute error in ATE estimation.

Outperforms neural networks with over 80% deviation and incorrect sign predictions.

Enables understanding of trade-offs in circuit design through explicit 'what-if' analysis.

Abstract

Analog-mixed-signal (AMS) circuits are highly non-linear and operate on continuous real-world signals, making them far more difficult to model with data-driven AI than digital blocks. To close the gap between structured design data (device dimensions, bias voltages, etc.) and real-world performance, we propose a causal-inference framework that first discovers a directed-acyclic graph (DAG) from SPICE simulation data and then quantifies parameter impact through Average Treatment Effect (ATE) estimation. The approach yields human-interpretable rankings of design knobs and explicit 'what-if' predictions, enabling designers to understand trade-offs in sizing and topology. We evaluate the pipeline on three operational-amplifier families (OTA, telescopic, and folded-cascode) implemented in TSMC 65nm and benchmark it against a baseline neural-network (NN) regressor. Across all circuits the causal model reproduces simulation-based ATEs with an average absolute error of less than 25%, whereas the neural network deviates by more than 80% and frequently predicts the wrong sign. These results demonstrate that causal AI provides both higher accuracy and explainability, paving the way for more efficient, trustworthy AMS design automation.