HeaRT: A Hierarchical Circuit Reasoning Tree-Based Agentic Framework for AMS Design Optimization

TL;DR

HeaRT is a hierarchical reasoning framework that enhances AMS design automation by improving accuracy and speed, demonstrating significant gains over baselines on a circuit benchmark.

Contribution

This work introduces HeaRT, a novel hierarchical circuit reasoning approach that improves automation accuracy and adaptability in AMS design optimization.

Findings

HeaRT improves F1(subcircuits) by >= 13.5% over baselines.

HeaRT improves F1(loops) by >= 37.8% over baselines.

HeaRT achieves >= 3x faster convergence in incremental tasks.

Abstract

Conventional AI-driven AMS design automation algorithms remain constrained by their reliance on high-quality datasets to capture underlying circuit behavior, coupled with poor transferability across architectures, and a lack of adaptive mechanisms. This work proposes HeaRT, a hierarchical circuit reasoning-based agentic framework for automation loops and a step toward adaptive, human-style design optimization. HeaRT consistently improves F1(subcircuits) by >= 13.5% and F1(loops) by >= 37.8% over few-shot prompting baselines across multiple LLM backbones on our 40-circuit AMS benchmark of flattened SPICE netlists, even as circuit complexity increases. Our experiments further show that HeaRT achieves >= 3x faster convergence in incremental design adaptation tasks under specification shifts across diverse optimization approaches, supporting both topology reconfiguration and sizing.