arXiv:2409.09376cs.LGstat.ML2024-09被引 4

用神经网络非迭代学习扩散模型中的最优传输路径。

BM$^2$: Coupled Schrödinger Bridge Matching

  • 通过耦合桥接匹配,直接学习目标分布间的动态传输映射。
  • 在可解析扩散过程下,实现比传统迭代方法更快的收敛速度。
  • 适合需要高效生成样本的扩散模型研究者使用。

Schrödinger桥通过参考过程建立两个目标分布之间的动态传输映射,同时求解相关的熵正则最优传输问题。本文考虑样本可得且参考扩散过程具有可解析动力学的情形,提出一种基于神经网络的耦合桥接匹配(BM²)方法,无需迭代即可学习Schrödinger桥。我们进行了初步的理论分析,证明了其收敛性,并通过数值实验验证了该方法的有效性。

原文摘要 · Abstract (English)

A Schrödinger bridge establishes a dynamic transport map between two target distributions via a reference process, simultaneously solving an associated entropic optimal transport problem. We consider the setting where samples from the target distributions are available, and the reference diffusion process admits tractable dynamics. We thus introduce Coupled Bridge Matching (BM$^2$), a simple non-iterative approach for learning Schrödinger bridges with neural networks. A preliminary theoretical analysis of the convergence properties of BM$^2$ is carried out, supported by numerical experiments that demonstrate the effectiveness of our proposal.

扩散模型最优传输神经网络

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。