通过流匹配的反事实生成：耦合敏感的端到端速率
Counterfactual Generation via Flow Matching: Coupling-Sensitive End-to-End Rates
- Johns Hopkins University（约翰霍普金斯大学）
- University of California, Davis（加州大学戴维斯分校）
- Amazon（亚马逊）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本文提出一种基于流匹配的反事实生成方法，结合双鲁棒目标与学习耦合，并给出耦合敏感的KL界和端到端保证，实验显示随机采样器在有限步数下优于确定性采样器。
AI中文摘要：
反事实生成旨在使用在事实分配机制下收集的观测数据，在假设干预或决策下对结果进行采样。我们开发了一种流匹配方法，将样本分割、双鲁棒训练目标与观测源结果和从拟合的条件结果模型中抽取的目标结果之间的学习耦合相结合。为了实现有限步生成，我们利用基于高斯平滑插值的分数校正随机采样器。我们的主要理论贡献是常数步欧拉离散化的耦合敏感KL界：误差由所选耦合下源-目标位移的矩控制，而不是由速度场的全局均匀正则性控制，并且对环境维度具有近线性依赖。我们还为从数据中估计条件结果模型和源-目标耦合时，学习的速度和分数场建立了有限样本非参数保证。这些界分离了近似、耦合替换、干扰估计、泛化和蒙特卡洛误差，并与采样器分析相结合，为反事实生成提供了端到端保证。在合成和半合成图像基准上的实验支持了耦合依赖理论，并表明在有限离散化预算下，随机采样器可以优于相应的确定性ODE采样器。
英文摘要：
Counterfactual generation seeks to sample outcomes under a hypothetical intervention or decision using observational data collected under the factual assignment mechanism. We develop a flow-matching approach that combines a sample-split, doubly robust training objective with a learned coupling between observed source outcomes and target outcomes drawn from a fitted conditional outcome model. To enable finite-step generation, we leverage a score-corrected stochastic sampler based on a Gaussian-smoothed interpolation. Our main theoretical contribution is a coupling-sensitive KL bound for constant-step Euler discretization: the error is controlled by moments of the source--target displacement under the chosen coupling, rather than by global uniform regularity of the velocity field, and has near-linear dependence on the ambient dimension. We also establish finite-sample non-parametric guarantees for the learned velocity and score fields when both the conditional outcome model and the source-target coupling are estimated from data. These bounds separate approximation, coupling-replacement, nuisance-estimation, generalization, and Monte Carlo errors and, combined with the sampler analysis, yield an end-to-end guarantee for counterfactual generation. Experiments on synthetic and semi-synthetic image benchmarks support the coupling-dependent theory and show that, at finite discretization budgets, the stochastic sampler can outperform the corresponding deterministic ODE sampler.
