Thread by @jeankaddour

Causal effect estimation is fundamental for decision-making in many domains, e.g., personalized medicine. Having access to observational data only, how can we infer causal effects of structured treatments, e.g., molecular graphs of drugs, images, texts, etc.? 1/4 #NeurIPS

We propose the generalized Robinson decomposition (GRD) for such treatments: it isolates the causal estimands (reducing regularization bias), allows one to plug in arbitrary models for learning, and possesses a quasi-oracle error bound under mild assumptions. 2/4

We implement an NN-based instance of GRD: Structured Intervention Networks. In experiments with molecular/small-world graphs, it outperforms prior approaches. There is plenty of future work to be done: datasets/applications, hidden confounding, consistency guarantees, ... 3/4

Joint work with @_zhuyuchen, Qi, Matt and Ricardo.

Video:

Paper: arxiv.org/pdf/2106.01939…

Code: github.com/JeanKaddour/SIN

4/4