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

2025-11-24

AI Summary

To address three key bottlenecks in analog/mixed-signal (AMS) circuit design automation—reliance on high-quality labeled data, poor cross-architecture transferability, and absence of adaptive mechanisms—this paper proposes a human-like hierarchical circuit reasoning framework. Our method introduces: (1) a novel proxy-based design optimization engine grounded in a hierarchical reasoning tree; (2) token-efficient inference integrated with multi-granularity behavioral circuit modeling; and (3) a closed-loop controlled design adaptation paradigm. Evaluated on a benchmark suite of 40 circuits, the framework achieves 97.2% inference accuracy and 98.5% Pass@1, while reducing timing overhead by 50% and accelerating convergence by 3.1×. It thus simultaneously delivers high accuracy, strong generalization across architectures, low latency, and faithful preservation of designer intent.

Technology Category

• Machine Learning: Hardware-aware ML
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

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 foundational reasoning engine for automation loops and a first step toward intelligent, adaptive, human-style design optimization. HeaRT consistently demonstrates reasoning accuracy >97% and Pass@1 performance >98% across our 40-circuit benchmark repository, even as circuit complexity increases, while operating at <0.5x real-time token budget of SOTA baselines. Our experiments show that HeaRT yields >3x faster convergence in both sizing and topology design adaptation tasks across diverse optimization approaches, while preserving prior design intent.

Problem

• Addresses limitations of AI-driven AMS design automation algorithms
• Improves transferability across circuit architectures and adaptivity
• Enhances reasoning accuracy and efficiency in design optimization

Innovation

• Hierarchical reasoning tree for circuit design optimization
• Adaptive human-style optimization without dataset dependency
• Real-time token efficiency with high reasoning accuracy