Logic Synthesis with Generative Deep Neural Networks

TL;DR

This paper introduces a novel logic synthesis rewriting operator using a generative deep neural network called Circuit Transformer, improving circuit transformation while maintaining equivalence, and demonstrating effectiveness on benchmark tests.

Contribution

The paper presents a new logic synthesis rewriting operator based on Circuit Transformer, with a specialized training scheme and integration with existing techniques for scalability.

Findings

Effective circuit rewriting demonstrated on IWLS 2023 benchmarks

Improved scalability through integration with state-of-the-art techniques

Self-improvement training enhances the model's optimality

Abstract

While deep learning has achieved significant success in various domains, its application to logic circuit design has been limited due to complex constraints and strict feasibility requirement. However, a recent generative deep neural model, "Circuit Transformer", has shown promise in this area by enabling equivalence-preserving circuit transformation on a small scale. In this paper, we introduce a logic synthesis rewriting operator based on the Circuit Transformer model, named "ctrw" (Circuit Transformer Rewriting), which incorporates the following techniques: (1) a two-stage training scheme for the Circuit Transformer tailored for logic synthesis, with iterative improvement of optimality through self-improvement training; (2) integration of the Circuit Transformer with state-of-the-art rewriting techniques to address scalability issues, allowing for guided DAG-aware rewriting. Experimental results on the IWLS 2023 contest benchmark demonstrate the effectiveness of our proposed rewriting methods.