用扩散路径优化反向KL，让神经采样器一步生成多模态数据。
Training Neural Samplers with Reverse Diffusive KL Divergence
- 在扩散轨迹上最小化反向KL，避免模式坍塌
- 单步采样即可逼近多模态目标分布
- 适合需要高效多模态采样的生成任务
训练生成模型从非归一化密度函数采样是机器学习中的重要挑战。传统方法依赖于反向Kullback-Leibler（KL）散度，因其可计算性，但其模式聚焦行为会阻碍对多模态目标分布的有效逼近。为此，我们提出沿模型与目标密度的扩散轨迹最小化反向KL，该目标称为反向扩散KL散度，使模型能捕捉多个模式。基于此目标，我们训练出能在一步内高效生成目标分布样本的神经采样器。实验表明，该方法在多种Boltzmann分布上均提升了采样性能，包括合成多模态密度和多体粒子系统。
原文摘要 · Abstract (English)
Training generative models to sample from unnormalized density functions is an important and challenging task in machine learning. Traditional training methods often rely on the reverse Kullback-Leibler (KL) divergence due to its tractability. However, the mode-seeking behavior of reverse KL hinders effective approximation of multi-modal target distributions. To address this, we propose to minimize the reverse KL along diffusion trajectories of both model and target densities. We refer to this objective as the reverse diffusive KL divergence, which allows the model to capture multiple modes. Leveraging this objective, we train neural samplers that can efficiently generate samples from the target distribution in one step. We demonstrate that our method enhances sampling performance across various Boltzmann distributions, including both synthetic multi-modal densities and n-body particle systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。