DeDiCa：面向隐藏混杂因素的离散因果生成模型
DeDiCa: A Scalable Deconfounding Discrete Causal Generative Model
arXiv 上这篇 DeDiCa 论文挺有意思，训练一次就能在毫秒内回答因果干预问题，还能在 CPU 上跑，比因果基础模型省时省内存。
DeDiCa 是一个针对离散变量与隐藏混杂因素的因果推断框架，通过对因果贝叶斯网络做张量化参数化实现可计算推断。模型只需在观测数据上用最大似然训练一次，对隐藏混杂因素做精确边缘化，之后即可估计任意可识别的干预查询。在 ATE 估计上，DeDiCa 与最先进的神经因果生成模型成绩相当，同时支持低内存的 CPU 训练和毫秒级精确干预推断。与因果基础模型相比，它随观测样本增加扩展性更好，推断时间和内存占用显著更低。
DeDiCa: A Scalable Deconfounding Discrete Causal Generative Model
We introduce DeDiCa, a practical framework for efficient causal inference with discrete variables and hidden confounding that revisits classical probabilistic modeling through a tensorized parameterization of causal Bayesian networks, enabling tractable inference. Given a causal graph, DeDiCa is trained once on observational data by maximum likelihood, exactly marginalizing over hidden confounders, and can then estimate any (proxy-)identifiable interventional query. Despite this deliberately simple parameterization, DeDiCa matches state-of-the-art neural causal generative models in ATE estimation, while supporting memory-efficient CPU training and exact interventional inference in milliseconds. Compared with causal foundation models, DeDiCa scales more effectively with additional observational samples and achieves competitive accuracy at substantially lower inference time and memory. DeDiCa provides a tractable, accurate, and scalable approach to causal inference in discrete settings with hidden confounding.
