Particle Guidance: non-I.I.D. Diverse Sampling with Diffusion Models
This addresses the issue of inefficient diverse sampling in generative models for applications like image and molecular generation, representing a novel method rather than an incremental improvement.
The paper tackles the problem of improving diversity and sample efficiency in generative models by moving beyond the independent sampling assumption, proposing particle guidance to enforce diversity in diffusion models. Empirically, it increases diversity without affecting quality in conditional image generation and reduces the state-of-the-art median error by 13% on average in molecular conformer generation.
In light of the widespread success of generative models, a significant amount of research has gone into speeding up their sampling time. However, generative models are often sampled multiple times to obtain a diverse set incurring a cost that is orthogonal to sampling time. We tackle the question of how to improve diversity and sample efficiency by moving beyond the common assumption of independent samples. We propose particle guidance, an extension of diffusion-based generative sampling where a joint-particle time-evolving potential enforces diversity. We analyze theoretically the joint distribution that particle guidance generates, how to learn a potential that achieves optimal diversity, and the connections with methods in other disciplines. Empirically, we test the framework both in the setting of conditional image generation, where we are able to increase diversity without affecting quality, and molecular conformer generation, where we reduce the state-of-the-art median error by 13% on average.