pith. sign in

The Annals of Statistics , pages=

3 Pith papers cite this work. Polarity classification is still indexing.

3 Pith papers citing it

years

2026 3

verdicts

UNVERDICTED 3

representative citing papers

Sinkhorn Treatment Effects: A Causal Optimal Transport Measure

stat.ML · 2026-05-08 · unverdicted · novelty 7.0

The Sinkhorn treatment effect is a new entropic optimal transport measure of divergence between counterfactual distributions that admits first- and second-order pathwise differentiability, debiased estimators, and asymptotically valid tests for distributional treatment effects.

Risk-Controlled Post-Processing of Decision Policies

stat.ML · 2026-05-07 · unverdicted · novelty 7.0

Risk-controlled post-processing yields a threshold-structured policy that follows the baseline except where an oracle fallback sharply reduces conditional violation risk, achieving O(log n/n) expected excess risk in i.i.d. settings and exact risk control under exchangeability.

citing papers explorer

Showing 3 of 3 citing papers.

  • Exploiting independence constraints for efficient estimation of bounds on causal effects in the presence of unmeasured confounding stat.ME · 2026-05-12 · unverdicted · none · ref 43

    An influence function projection approach exploits graph-implied conditional independences to improve the efficiency of semiparametric estimators for upper and lower bounds on average causal effects under sensitivity models for unmeasured confounding.

  • Sinkhorn Treatment Effects: A Causal Optimal Transport Measure stat.ML · 2026-05-08 · unverdicted · none · ref 174

    The Sinkhorn treatment effect is a new entropic optimal transport measure of divergence between counterfactual distributions that admits first- and second-order pathwise differentiability, debiased estimators, and asymptotically valid tests for distributional treatment effects.

  • Risk-Controlled Post-Processing of Decision Policies stat.ML · 2026-05-07 · unverdicted · none · ref 117

    Risk-controlled post-processing yields a threshold-structured policy that follows the baseline except where an oracle fallback sharply reduces conditional violation risk, achieving O(log n/n) expected excess risk in i.i.d. settings and exact risk control under exchangeability.