The optimal bid in a second-price ad auction equals the conditional average treatment effect of the ad, and the paper's Thompson-sampling algorithm learns that bid and therefore the ad effect while reducing experimentation cost.
B.2 Highest competing bid for SPAs Even though we maintain the assumption of treatment exogeneity, we still have to ac- count for censoring of the highest competing bid
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Online Causal Inference for Advertising in Real-Time Bidding Auctions
The optimal bid in a second-price ad auction equals the conditional average treatment effect of the ad, and the paper's Thompson-sampling algorithm learns that bid and therefore the ad effect while reducing experimentation cost.