AI 中文总结
该研究针对纵向观察数据反事实估计的互信息冲突，提出CSSD和CSSPD两个因果状态空间模型，在MIMIC-III及癌症模拟数据集上均优于基线模型，为因果推断提供了新方案。
AI 中文摘要
从纵向观察数据中估计随时间变化的反事实结果是临床决策支持的核心。现有方法依赖领域混淆——使表示对处理分配不变的对抗性训练——但这种不变性产生了互信息冲突：它抑制了准确结果预测所需的与处理相关的协变量信号。我们通过反事实预测误差的Jensen-Shannon散度界形式化了这种张力，并开发了两个互补模型。CSSD（带直接解码器的因果状态空间模型）采用选择性状态空间模型，搭配并行多步解码器，通过单次前向传播生成所有预测 horizon，消除了累积的 rollout 误差。CSSPD（带预测正则化和直接解码器的因果状态空间模型）在CSSD基础上加入对比预测编码和局部信息最大化，以增强平衡表示中的时间可预测性，并恢复被领域混淆破坏的局部协变量信息。在MIMIC-III数据集上，CSSPD在所有horizon τ≥2时的反事实RMSE均低于Causal Transformer，编码器成本为O(T)，增益从0.02（2步）到0.07（6步）不等。在跨混淆强度γ∈{0,1,2,3,4}的癌症模拟中，CSSPD在γ≤3时优于CT，幅度为25.9%至37.0%，而CSSD实现了最低的总体平均RMSE（较CT降低12.7%），证实了互信息冲突分析。据我们所知，这是首个将平衡-预测互信息冲突形式化，并通过互补的预测和信息论训练目标提出结构化解决方案的工作。
英文摘要
Estimating counterfactual outcomes over time from longitudinal observational data is central to clinical decision support. Existing methods rely on domain confusion -- adversarial training that renders representations invariant to treatment assignment -- yet this invariance creates a mutual information conflict: it suppresses treatment-correlated covariate signals necessary for accurate outcome prediction. We formalise this tension via a Jensen-Shannon divergence bound on counterfactual prediction error and develop two complementary models. CSSD (Causal State-Space model with Direct decoder) adapts selective State Space Models with a parallel multi-step decoder that eliminates accumulated rollout error by producing all prediction horizons simultaneously in a single forward pass. CSSPD (Causal State-Space model with Predictive regularisation and Direct decoder) augments CSSD with Contrastive Predictive Coding and Local Information Maximisation to reinforce temporal predictability in the balancing representation and recover local covariate information destroyed by domain confusion. On MIMIC-III, CSSPD achieves lower counterfactual RMSE than the Causal Transformer at every horizon tau >= 2 at O(T) encoder cost, with gains from 0.02 (2-step) to 0.07 (6-step). On Cancer Simulation across confounding strengths gamma in {0,1,2,3,4}, CSSPD outperforms CT at gamma <= 3 (margins 25.9%--37.0%), and CSSD achieves the lowest overall average RMSE (12.7% reduction over CT), confirming the MI conflict analysis. To our knowledge, this is the first work to formalise the balancing-prediction MI conflict and propose a structured resolution through complementary predictive and information-theoretic training objectives.
CommentsSubmitted to AAAI 2027. 13 pages, 5 tables, appendices A-E
