Search papers, labs, and topics across Lattice.
3
0
2
1
Overlapping actions in A/B tests inflate noise and confidence intervals, but a novel meta-policy approach can significantly reduce this variance and enhance treatment effect estimation.
Online A/B testing's classic Difference-in-Means estimator is just off-policy Inverse Propensity Scoring in disguise.
Forget SNIPS: optimal additive baselines provably crush self-normalization for off-policy evaluation.
