Don't Play Favorites: Minority Guidance for Diffusion Models

by   Soobin Um, et al.

We explore the problem of generating minority samples using diffusion models. The minority samples are instances that lie on low-density regions of a data manifold. Generating sufficient numbers of such minority instances is important, since they often contain some unique attributes of the data. However, the conventional generation process of the diffusion models mostly yields majority samples (that lie on high-density regions of the manifold) due to their high likelihoods, making themselves highly ineffective and time-consuming for the task. In this work, we present a novel framework that can make the generation process of the diffusion models focus on the minority samples. We first provide a new insight on the majority-focused nature of the diffusion models: they denoise in favor of the majority samples. The observation motivates us to introduce a metric that describes the uniqueness of a given sample. To address the inherent preference of the diffusion models w.r.t. the majority samples, we further develop minority guidance, a sampling technique that can guide the generation process toward regions with desired likelihood levels. Experiments on benchmark real datasets demonstrate that our minority guidance can greatly improve the capability of generating the low-likelihood minority samples over existing generative frameworks including the standard diffusion sampler.


page 7

page 12

page 13

page 16

page 17

page 19

page 20

page 21


Generating High Fidelity Data from Low-density Regions using Diffusion Models

Our work focuses on addressing sample deficiency from low-density region...

Hierarchically branched diffusion models for efficient and interpretable multi-class conditional generation

Diffusion models have achieved justifiable popularity by attaining state...

Learning Energy-Based Models by Cooperative Diffusion Recovery Likelihood

Training energy-based models (EBMs) with maximum likelihood estimation o...

Diffusion map particle systems for generative modeling

We propose a novel diffusion map particle system (DMPS) for generative m...

Manifold-Guided Sampling in Diffusion Models for Unbiased Image Generation

Diffusion models are a powerful class of generative models that can prod...

Don't be so negative! Score-based Generative Modeling with Oracle-assisted Guidance

The maximum likelihood principle advocates parameter estimation via opti...

Refining Generative Process with Discriminator Guidance in Score-based Diffusion Models

While the success of diffusion models has been witnessed in various doma...

Please sign up or login with your details

Forgot password? Click here to reset