Flow-based Generative Modeling of Potential Outcomes and Counterfactuals
基于流的潜在结果和反事实建模
机构 * School of Industrial and Systems Engineering(工业与系统工程学院) ; Georgia Institute of Technology(佐治亚理工学院) ; Elmore Family School of Electrical and Computer Engineering(埃洛姆家族电气与计算机工程学院) ; Purdue University(普渡大学)
AI总结 本文提出PO-Flow框架,通过连续归一化流模型联合建模潜在结果分布和事实条件下的反事实结果,实现个性化预测和因果推断。
Comments Accepted at 2026 IEEE International Symposium on Information Theory (ISIT 2026)