TAG: Learning Circuit Spatial Embedding From Layouts

TL;DR

This paper introduces TAG, a novel machine learning framework that learns spatial embeddings of AMS circuit layouts using text, self-attention, and graph techniques, enabling improved prediction tasks without manual labeling.

Contribution

The paper presents a new circuit embedding method that leverages layout data with self-attention and text embedding, enhancing automation in AMS circuit design.

Findings

Successfully predicts layout distances with industrial benchmarks.

Demonstrates transferability to layout matching, wirelength, and parasitic capacitance tasks.

Reduces reliance on manual labeling in circuit representation learning.

Abstract

Analog and mixed-signal (AMS) circuit designs still rely on human design expertise. Machine learning has been assisting circuit design automation by replacing human experience with artificial intelligence. This paper presents TAG, a new paradigm of learning the circuit representation from layouts leveraging text, self-attention and graph. The embedding network model learns spatial information without manual labeling. We introduce text embedding and a self-attention mechanism to AMS circuit learning. Experimental results demonstrate the ability to predict layout distances between instances with industrial FinFET technology benchmarks. The effectiveness of the circuit representation is verified by showing the transferability to three other learning tasks with limited data in the case studies: layout matching prediction, wirelength estimation, and net parasitic capacitance prediction.