PROUD: PaRetO-gUided Diffusion Model for Multi-objective Generation
This work addresses the challenge of multi-objective generation for applications like image and protein design, offering a method to balance trade-offs without quality loss, though it is incremental in improving existing diffusion models.
The paper tackles the problem of generating samples that satisfy multiple conflicting properties without compromising quality, by formulating a constrained optimization to ensure Pareto optimality and introducing PROUD, a diffusion model that dynamically adjusts gradients, achieving superior generation quality and approaching Pareto optimality in image and protein generation tasks.
Recent advancements in the realm of deep generative models focus on generating samples that satisfy multiple desired properties. However, prevalent approaches optimize these property functions independently, thus omitting the trade-offs among them. In addition, the property optimization is often improperly integrated into the generative models, resulting in an unnecessary compromise on generation quality (i.e., the quality of generated samples). To address these issues, we formulate a constrained optimization problem. It seeks to optimize generation quality while ensuring that generated samples reside at the Pareto front of multiple property objectives. Such a formulation enables the generation of samples that cannot be further improved simultaneously on the conflicting property functions and preserves good quality of generated samples. Building upon this formulation, we introduce the PaRetO-gUided Diffusion model (PROUD), wherein the gradients in the denoising process are dynamically adjusted to enhance generation quality while the generated samples adhere to Pareto optimality. Experimental evaluations on image generation and protein generation tasks demonstrate that our PROUD consistently maintains superior generation quality while approaching Pareto optimality across multiple property functions compared to various baselines.