2 papers
cs.LG2026
Guide, Not Bind: Why Defeasible Priors Fail in Augmented Lagrangian Causal Discovery
Sairam Sundararaman, Sara Girdhar, Manit Narasimha Murthy +2
Differentiable causal discovery methods increasingly encode expert priors as forbidden-edge constraints enforced by an Augmented Lagrangian (ALM) penalty, on the assumption that a…
cs.LG2025
The Final-Stage Bottleneck: A Systematic Dissection of the R-Learner for Network Causal Inference
S Sairam, Sara Girdhar, Shivam Soni
The R-Learner is a powerful, theoretically-grounded framework for estimating heterogeneous treatment effects, prized for its robustness to nuisance model errors. However, its appli…