Inverse design with conditional cascaded diffusion models

TL;DR

This paper introduces a conditional cascaded diffusion model (cCDM) for inverse design tasks, demonstrating its advantages over cGANs in capturing details and constraints when sufficient high-resolution training data is available.

Contribution

The paper proposes the cCDM framework for multi-resolution inverse design, showing its stability, independent training of models, and superior performance with ample high-resolution data.

Findings

cCDM outperforms cGAN in detail capture and constraint preservation with enough data

Performance of cCDM declines with limited data, losing superiority to cGAN

Diffusion models excel in pixel-wise accuracy but not always in optimality or constraint satisfaction

Abstract

Adjoint-based design optimizations are usually computationally expensive and those costs scale with resolution. To address this, researchers have proposed machine learning approaches for inverse design that can predict higher-resolution solutions from lower cost/resolution ones. Due to the recent success of diffusion models over traditional generative models, we extend the use of diffusion models for multi-resolution tasks by proposing the conditional cascaded diffusion model (cCDM). Compared to GANs, cCDM is more stable to train, and each diffusion model within the cCDM can be trained independently, thus each model's parameters can be tuned separately to maximize the performance of the pipeline. Our study compares cCDM against a cGAN model with transfer learning. Our results demonstrate that the cCDM excels in capturing finer details, preserving volume fraction constraints, and minimizing compliance errors in multi-resolution tasks when a sufficient amount of high-resolution training data (more than 102 designs) is available. Furthermore, we explore the impact of training data size on the performance of both models. While both models show decreased performance with reduced high-resolution training data, the cCDM loses its superiority to the cGAN model with transfer learning when training data is limited (less than 102), and we show the break-even point for this transition. Also, we highlight that while the diffusion model may achieve better pixel-wise performance in both low-resolution and high-resolution scenarios, this does not necessarily guarantee that the model produces optimal compliance error or constraint satisfaction.