TS-PostDiff adaptively blends Thompson Sampling and uniform randomization using the posterior probability of small arm differences to improve the tradeoff between reward and statistical inference in two-armed experiments.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2021 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Algorithms for Adaptive Experiments that Trade-off Statistical Analysis with Reward: Combining Uniform Random Assignment and Reward Maximization
TS-PostDiff adaptively blends Thompson Sampling and uniform randomization using the posterior probability of small arm differences to improve the tradeoff between reward and statistical inference in two-armed experiments.