A robust design selector minimizes worst-case planning risk over an ambiguity set of exposure mechanisms, with Wasserstein bounds and selector theorems, yielding different recommendations on public datasets.
Limiting bias from test-control interference in online marketplace experiments
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
stat.ML 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
A support-aware offline decision framework for reserve-policy selection that outputs certified policies and shortlists instead of rankings, with a finite-catalog guarantee preserving the best supported policy.
citing papers explorer
-
Choosing Online Experiment Designs under Interference in Ads, Recommendations, and Member-Experience Systems
A robust design selector minimizes worst-case planning risk over an ambiguity set of exposure mechanisms, with Wasserstein bounds and selector theorems, yielding different recommendations on public datasets.
-
Support-aware offline policy selection for advertising marketplaces
A support-aware offline decision framework for reserve-policy selection that outputs certified policies and shortlists instead of rankings, with a finite-catalog guarantee preserving the best supported policy.