AC-Refiner: Efficient Arithmetic Circuit Optimization Using Conditional Diffusion Models
This work addresses the problem of efficient arithmetic circuit optimization for digital system designers, offering a novel deep learning-based method that improves upon existing approaches.
The paper tackles the challenge of optimizing arithmetic circuits like adders and multipliers by proposing AC-Refiner, a framework that uses conditional diffusion models to reframe circuit synthesis as a conditional image generation task, resulting in designs with superior Pareto optimality that outperform state-of-the-art baselines.
Arithmetic circuits, such as adders and multipliers, are fundamental components of digital systems, directly impacting the performance, power efficiency, and area footprint. However, optimizing these circuits remains challenging due to the vast design space and complex physical constraints. While recent deep learning-based approaches have shown promise, they struggle to consistently explore high-potential design variants, limiting their optimization efficiency. To address this challenge, we propose AC-Refiner, a novel arithmetic circuit optimization framework leveraging conditional diffusion models. Our key insight is to reframe arithmetic circuit synthesis as a conditional image generation task. By carefully conditioning the denoising diffusion process on target quality-of-results (QoRs), AC-Refiner consistently produces high-quality circuit designs. Furthermore, the explored designs are used to fine-tune the diffusion model, which focuses the exploration near the Pareto frontier. Experimental results demonstrate that AC-Refiner generates designs with superior Pareto optimality, outperforming state-of-the-art baselines. The performance gain is further validated by integrating AC-Refiner into practical applications.